Graduate Theses and Dissertations
Graduate College
2013
Predictors of gains in inductive reasoning strategies
and everyday functioning: Results from the
Advanced Cognitive Training for Independent and
Vital Elderly (ACTIVE) Study
Joan Blaser Baenziger
Iowa State University
Follow this and additional works at: http://lib.dr.iastate.edu/etd
Part of the Quantitative, Qualitative, Comparative, and Historical Methodologies Commons, and
the Social Psychology Commons
Recommended Citation
Baenziger, Joan Blaser, "Predictors of gains in inductive reasoning strategies and everyday functioning: Results from the Advanced
Cognitive Training for Independent and Vital Elderly (ACTIVE) Study" (2013). Graduate Theses and Dissertations. Paper 13420.
This Dissertation is brought to you for free and open access by the Graduate College at Digital Repository @ Iowa State University. It has been accepted
for inclusion in Graduate Theses and Dissertations by an authorized administrator of Digital Repository @ Iowa State University. For more
information, please contact [email protected].
Predictors of gains in inductive reasoning strategies and everyday functioning:
Results from the Advanced Cognitive Training for Independent and
Vital Elderly (ACTIVE) Study
by
Joan Blaser Baenziger
A dissertation submitted to the graduate faculty
in partial fulfillment of the requirements for the degree of
DOCTOR OF PHILOSOPHY
Major: Human Development and Family Studies
Program of Study Committee:
Jennifer Margrett, Co-Major Professor
Daniel W. Russell, Co-Major Professor
Carolyn Cutrona
Jan Melby
Mack Shelley, III
Iowa State University
Ames, Iowa
2013
Copyright © Joan Blaser Baenziger, 2013. All rights reserved.
ii
TABLE OF CONTENTS
LIST OF FIGURES ............................................................................................................. iv
LIST OF TABLES ................................................................................................................v
ACKNOWLEDGEMENTS ................................................................................................. vi
ABSTRACT ....................................................................................................................... vii
CHAPTER 1. INTRODUCTION ..........................................................................................1
CHAPTER 2. LITERATURE REVIEW ................................................................................4
Cognitive Change and Decline ...................................................................................4
Prevalence Rates of Cognitive Decline.......................................................................5
Compression of Cognitive Morbidity .........................................................................7
Theories of Aging ......................................................................................................7
Theories of Cognition ................................................................................................9
Cognitive Decline with Age ..................................................................................... 10
Reasoning ................................................................................................................ 11
Consequences of Declines in Reasoning .................................................................. 12
Assessment of Reasoning Using Measures of Everyday Functioning ....................... 12
Cognitive Training Interventions (CTIs) .................................................................. 13
Cognitive Strategy Training Interventions (CSTIs) .................................................. 15
Research Questions .................................................................................................. 19
CHAPTER 3. METHODS ................................................................................................... 20
Study Design ........................................................................................................... 20
Participants and Procedures ..................................................................................... 21
Participant Eligibility ................................................................................... 21
Differences Between Participants and Nonparticipants ................................. 21
Interviewer Training ..................................................................................... 22
Pre-Intervention Assessment ........................................................................ 22
Reasoning Strategy Training ........................................................................ 22
Attrition ....................................................................................................... 25
Measures ................................................................................................................. 25
Demographic Variables ................................................................................ 25
Measures of Inductive Reasoning ................................................................. 25
Reasoning Strategies .................................................................................... 27
Everyday Functioning .................................................................................. 28
CHAPTER 4. RESULTS ..................................................................................................... 29
Sample Characteristics ............................................................................................. 29
Prediction of Strategy Use ....................................................................................... 29
Letter Series ............................................................................................................. 30
Letter Sets................................................................................................................ 32
Word Series ............................................................................................................. 34
Summary...................................................................................................... 35
iii
Prediction of Functioning Over Time ....................................................................... 36
Instrumental Activities of Daily Living ........................................................ 36
Observed Tasks of Daily Living ................................................................... 39
Summary...................................................................................................... 41
CHAPTER 5. DISCUSSION AND CONCLUSIONS.......................................................... 42
Impact of the Reasoning Strategy Training on Strategy Use ......................... 42
Effect of the Reasoning Strategy Training on Functional Status .................... 43
Implications for Reasoning Strategy training ................................................ 44
Screening Protocols for Recruitment of Participants ..................................... 47
Influence of Reasoning Strategies on Everyday Functioning......................... 47
Limitations .............................................................................................................. 49
Future Research ....................................................................................................... 50
Conclusions ............................................................................................................. 51
FIGURES ............................................................................................................................ 53
TABLES ............................................................................................................................. 60
APPENDIX A. INSTITUTIONAL REVIEW BOARD APPROVAL .................................. 73
APPENDIX B. FOUR REASONING TRAINING STRATEGIES ...................................... 75
REFERENCES.................................................................................................................... 76
iv
LIST OF FIGURES
Figure 1. Population age 65 and over and age 85 and over for selected years
1900 - 2008 and projected from 2010 – 2050. U.S. Census Bureau
Decennial Census, 2010 ………………………………………………………………… 53
Figure 2. Life expectancy from 1900 - 2006 at ages 65 and 85, by sex. Federal
Interagency Forum on Aging-Related Statistics, 2007……………………………… 54
Figure 3. Impact of observed strategy gains due to reasoning training on everyday
functioning .......................................................................................................... 55
Figure 4. Instrumental Activities of Daily Living over time ................................................ 56
Figure 5. Causal model for the Instrumental Activities of Daily Living (IADL) .................. 57
Figure 6. Observed Tasks of Daily Living measured over time ........................................... 58
Figure 7. Causal model for the Observed Tasks of Daily Living (OTDL) ........................... 59
v
LIST OF TABLES
Table 1. Demographic Characteristics of ACTIVE Participants by Intervention Site .......... 60
Table 2. Correlations among Variables............................................................................... 61
Table 3. Hierarchical Linear Regression Results for Letter Series Strategy Use .................. 62
Table 4. Hierarchical Linear Regression Results for Letter Sets Strategy Use ..................... 63
Table 5. Hierarchical Linear Regression Results for Word Sets Strategy Use ..................... 64
Table 6. Correlations among the Instrumental Activities of Daily Living (IADL)
Model Variables ................................................................................................... 65
Table 7. Growth Curve Modeling Results for the Instrumental Activities of Daily
Living Intercept .................................................................................................... 66
Table 8. Growth Curve Modeling Results for the Instrumental Activities of Daily
Living Linear Term .............................................................................................. 67
Table 9 Growth Curve Modeling Results for the Instrumental Activities of Daily
Living Quadratic Term.......................................................................................... 68
Table 10. Correlations among the Observed Tasks of Daily Living (OTDL) Model
Variables .............................................................................................................. 69
Table 11. Growth Curve Modeling Results for the Observed Tasks of Daily Living
Intercept ............................................................................................................... 70
Table 12. Growth Curve Modeling Results for the Observed Tasks of Daily Living
Linear Term ......................................................................................................... 71
Table 13. Growth Curve Modeling Results for the Observed Tasks of Daily Living
Quadratic Term .................................................................................................... 72
vi
ACKNOWLEDGEMENTS
First, I wish to express my deepest appreciation to my major professors and my
mentors, Dr. Daniel W. Russell and Dr. Jennifer Margrett. They provided never ending
support and encouragement while I worked both on my graduate studies and my dissertation.
Specifically, Dr. Russell gave me excellent statistical advice and emotional support while Dr.
Margrett gave me her experience in cognitive aging studies and encouragement to “keep on
swimming” and never give up. Without their involvement and support I might not have
accomplished my goal of completing my dissertation. Second, I would like to express my
appreciation to my other committee members: Dr. Mack Shelley, Dr. Carolyn Cutrona, and
Dr. Jan Melby. Their comments and support were appreciated for their insights and wisdom.
Dr. Cutrona was there supporting me when I needed it the most and Dr. Shelley’s advice and
cheerful continence keep me continuing on my path. Dr. Melby’s kind words of
encouragement and knowledge of observational data was extremely valuable. Third, I would
like to thank my fellow students and friends for their friendship and “never give up spirit”
through thick and thin: Grace, Aradhana, Jinmyoung, and Mindy. Fourth, I would like to
thank: my children (Bill, Nick, Daniel, and Chris Paul), my daughter-in-laws (Sue and Sara)
and my many grandchildren who remind me they are watching and modeling older adults. To
both my brothers, Steve and Glenn, and to my sister-in-laws, Barb and Laura, as well as my
step-mother, Jean, my Dad and my Mom, I thank them for their love and faith in my
perseverance and capabilities. They always encouraged me to follow my dreams and to
believe in myself. Finally, I would like to express my greatest thanks to my husband,
Gregory Paul, who supported me with his love and encouragement day in and day out
through the whole process. I would like to say to him, “thank you for being there for me.”
vii
ABSTRACT
Prior research demonstrates that some cognitive abilities (i.e., memory, speed of processing
and reasoning) decline starting in the sixth decade of life. One mechanism underlying
training interventions is cognitive strategies which can maintain or enhance abilities and
associated everyday functioning. The present study focused on two research questions. First,
does participation in reasoning strategy training lead to increased use of the strategies in
performing reasoning tasks? Second, does participation in reasoning strategy training
influence subsequent changes in indicators of everyday functioning over a five-year period?
To address these issues data were analyzed from the Advanced Cognitive Training for
Independent and Vital Elderly (ACTIVE) study, a large 10-year investigation of the effects
of teaching cognitive strategies to a healthy sample of older adults (Jobe et al., 2001). A total
of 601 participants from that study who either received the reasoning training or were
assigned to a no-treatment control group were included in the analyses. Regarding the first
research question, hierarchical regression analyses indicated that the reasoning training was
very effective in enhancing the use of strategies by participants. Additional analyses found
that the intervention was most effective for participants who were younger, better educated,
and White although all groups benefited from training. Regarding the second research
question, the results of growth curve modeling analyses indicated receipt of the reasoning
strategy training was not related to change in functioning over the five-year period among
participants. These results indicate that, although the intervention was effective in enhancing
the use of reasoning strategies, these changes did not generalize to everyday functioning
among this sample of older adults. Implications of these results for enhancing the cognitive
abilities of older adults and improving their functional status are discussed.
1
CHAPTER 1. INTRODUCTION
The 65 and older population increased at a faster rate (15.1%) than the general
population (9.7%) between 2000 and 2010. This age group represents 12.9% of the total
U.S. population or approximately one in every eight Americans (National Institute on Aging,
2012; U.S. Department of Health and Human Services, 2011). This figure is projected to
grow to 19% by 2030 or approximately one in every five Americans (U.S. Department of
Health and Human Services, 2011). By 2050 the number of Americans aged 65 and older is
projected to be 88.5 million which is more than twice the population of 40.2 million in 2010
(Federal Interagency Forum on Aging Related Statistics, 2007). Persons reaching age 65
today have an average life expectancy of an additional 18.6 years. As can be seen in Figure 1,
there is a dramatic rise in the number of persons age 65 and older as compared to the
previous century.
As shown in Figure 2 there are differences in the life expectancy of older adults by
sex. On average females live an additional 19.9 years past age 65 whereas males live an
additional 17.2 years (Federal Interagency Forum on Aging Related Statistics, 2007). The
number of males has increased disproportionately when compared to the female population,
resulting in a narrowing of the ratio of males to females in this age group which is expected
to impact the cost of Medicare and Medicaid (U.S. Census Bureau, 2010).
Glen Elder, a prominent researcher in the field of aging, coined the phrase “across the
life course” to refer to non-chronological aging (Elder, 1992) and drawing attention to the
fact that aging is a developmental, lifelong process. That is, individuals develop not only
from birth to young adulthood, but continue to develop over their individual life course.
2
Their unique life course is viewed as the summation of all their experiences during their
lifetime and this process continues until death.
One important issue concerning older adults is their cognitive health and how their
cognitive health affects their everyday functioning. There is an extensive research literature
related to cognitive functioning, cognitive maintenance, and cognitive change. One of these
important abilities is reasoning. A decline of cognitive reasoning often leads to loss of
independence for older adults and results in either formal or informal caregiving or
residential reassignment.
The possibility of maintaining the ability to reason or to delay the onset of the decline
in reasoning ability is therefore of great interest to both researchers and older adults. This
interest has led to research on teaching reasoning strategies which could help older adults
maintain or enhance their reasoning abilities and thereby delay the onset of cognitive decline.
Only a few studies have tried to intervene with older adults by teaching cognitive strategies
to determine if increased strategy use can result in better outcomes in their daily lives.
The Advanced Cognitive Training for Independent and Vital Elderly (ACTIVE) is
one such study. Participants in the ACTIVE study were trained in reasoning and other
cognitive strategies over a 5-6 week period of time and then participants were followed over
the subsequent 10 years to evaluate long term effects of the intervention. The current
investigation conducted secondary analyses of data from the ACTIVE study for individuals
who received training in the use of four inductive reasoning strategies. These strategies are
unique techniques taught to help participants discern patterns in typical information they find
in their everyday lives. For example, older adults would apply these strategies on a daily
basis when attempting to find the correct time or place to catch a bus/train when leaving their
3
homes. Finding the correct time or place allows older adults to reach their important
scheduled appointments (i.e., the doctor’s or dentist’s office) or to attend a fitness class, meet
a friend, or shop for groceries. Another typical example would be when an older adult needs
to take their next medication. This would require them to read often complex instructions on
a medication bottle. Learning reasoning strategies to accomplish such everyday tasks helps
older adults function successfully and increases the chance that they will maintain
independent lives for a longer period of time. Therefore, inductive reasoning clearly is a
vital cognitive ability which needs to be maintained or enhanced as it slowly declines during
the sixth decade of life.
In order to further study the impact of strategy training on older adults, secondary
data were analyzed using participants from both the reasoning and control groups of the
ACTIVE study. Two research questions were addressed. First, did participation in the
reasoning strategy training lead to increased use of the strategies after training? Second, did
participation in the reasoning strategy training influence changes in everyday functioning
status over time (i.e., prior to the intervention to 5 years following the intervention)?
4
CHAPTER 2. LITERATURE REVIEW
As individuals age, cognitive capacities and capabilities develop over time.
Cognition and cognitive functioning can occur in both normative and non-normative ways
that affect the cognitive health of older adults. “Cognition” refers to a group of processes
occurring in the brain whereas “cognitive functioning” refers to “performance-based
indicators of cognitive ability or skill” (Dixon, Backman, & Lars-Göran, 2004, p. 3-4).
Cognitive impairment, however, is a non-normative aging-related process and includes both
memory and non-memory deficits. It involves a group of symptoms currently thought of as a
transition phase between healthy cognitive aging and dementia (DeCarli, 2003). Cognition,
like physical health, can be viewed along a continuum - from optimal functioning to mild
cognitive impairment to severe dementia (U.S. Department of Health and Human Services,
2007).
Cognitive Change and Decline
Cognitive change and decline in some cognitive abilities during older adulthood is
normal. However, cognitive impairment is not normal and can result from a variety of
causes (i.e., disease, trauma). It is an older adult’s inability to perform cognitively
demanding tasks (e.g., driving, taking medications, and managing finances) due to cognitive
impairment that frequently motivates them or others (i.e., spouses and adult children) to seek
assessment and diagnosis. Studies have reported that Instrumental Activities of Daily Living
(IADLs), such as answering the phone and handing finances (Fillenbaum, 1987a,b;
Fillenbaum & Smyer, 1981; Lawton & Brody, 1969), decline before reductions in the ability
to perform activities found on the Activities of Daily Living scale (ADLs) such as grooming,
toileting, dressing, and eating (Fillenbaum & Smyer, 1981; Katz et al., 1963). That is,
5
activities requiring higher order cognitive functioning are “lost” before other, more basic
activities (Ashford, Hsu, Becker, Kuman, & Bekian, 1986; Reisberg, Ferris, de Leon, &
Crook, 1982). More recent research has revealed relatively modest declines in the
performance of cognitively complex everyday tasks for adults during their 60s, but steeper
patterns of decline in their late 70s and 80s (Schaie, 1996).
Not all cognitive abilities decline at the same rate. For example, working memory
and spatial orientation appear to decline before vocabulary. Speed of processing tends to
decline before abstract reasoning and working memory appears to decline before both speed
of processing and abstract reasoning (Ball et al., 2002; Salthouse, 1993). However, it is
important to note that these declines vary from individual to individual which leads to large
within-group variability in decline (Willis, Jay, Diehl, & Marsiske, 1992; Diehl, Willis, &
Schaie, 1995; Willis et al., 2006). Previous research suggests that some individuals are at
cognitive risk due to social and/or cultural differences in our society. These social and/or
cultural disadvantages (i.e., lower socio-economic status, lower education, and poorer health
access) can result in declines in different cognitive abilities (Schaie, 1999).
Prevalence Rates of Cognitive Decline
Cognitive decline at the individual level before age 60 is not normative (Siegler,
Hooker, Bosworth, Elias, & Spiro, 2010). The prevalence of cognitive impairment among
older adults age 65 and older without dementia ranges from 5% to 29%, but research findings
also indicate that the majority of older adults will not experience moderate to severe memory
impairment in their lifetime (Petersen et al., 1999). The prevalence of cognitive impairment
appears to increase steeply with advanced age (Heeringa et al., 2007; Plassman et al., 2007,
2008). The Aging, Demographics, and Memory Study (ADAMS), the first population-based
6
study which included individuals from all regions of the country to increase the knowledge of
cognitive impairment in the United States (Heeringa et al., 2007; Plassman et al., 2007). This
study began in the early 1990s and collected data from a subset of participants from the
larger, national Health and Retirement Study. Their analyses of the ADAMs data revealed
that in the 71- to 79-year-old age group, 16% showed evidence of cognitive impairment
without dementia whereas an additional 5% suffered from dementia.
More recent analyses comparing different decades of data on the prevalence of
cognitive impairment reveal that these rates have declined. For example, an analysis of the
Health and Retirement Study revealed a significant decline in the prevalence of severe
cognitive impairment among people aged 70 and older from 6% in 1993 to less than 4% in
1998 (Freedman, Martin, & Schoeni, 2002). This suggests there may be a slowing in the
proportion of older adults who develop severe cognitive impairment due to factors such as
increases in levels of education and changes in lifestyle. Despite a possible decrease in
severe CI, the increase in the number of older adults makes cognitive impairment an
important issue for the United States due to the large number of older adults with mild
cognitive impairment. Individuals with mild cognitive impairment experience the loss of
abilities such as making monetary transactions when shopping, the ability to drive, the ability
to do their taxes, and other abilities associated with daily living. A recent report by the
Centers for Disease Control and the Merck Company Foundation (2007) identifies the
prevention of cognitive decline as a key area where public health interventions can make
significant improvements in the quality of life of older adults.
7
Compression of Cognitive Morbidity
A change in the prevalence rate of cognitive impairment has been noted by recent
investigators using data from the Health and Retirement Study (Lang, Rieckmann, & Baltes,
2002). This change is known as “compression of morbidity” due to cognitive impairment
occurring much later in the course of the individual’s life. Compression of morbidity results
in a more severe decline in cognitive impairment closer to death (Langa et al., 2007).
Several factors during the past 15 years, such as better medical care and better health
behaviors, may have had an impact by reducing mortality (Crimmins et al., 2010). In
addition, older adults in more recent cohorts have received higher levels of education. The
proportion of adults aged 65 or older with a high school diploma increased from 53% in 1990
to 72% in 2003, whereas the proportion with a college degree increased from 11% to 17%
during this same time period (National Institutes of Health, 2011). Better health behaviors as
well as more wealth and social opportunities are associated with delays the decline of
cognitive abilities (i.e., compression of morbidity) compared to earlier cohorts (Ferri et al.,
2005; Plassman et al., 2006). Finally, greater wealth is associated with lower levels of
disability throughout the life course (Breitner et al., 1999; Callahan, Hendrie, & Tierney,
1995).
Theories of Aging
Life Span Theory (LST) deals with individual development from conception to old
age (Baltes, 1987; Baltes & Baltes, 1997). This theory originally consisted of three
components. First, development occurs throughout life. Second, older adults view life as a
series of gains and losses. These gains and losses can occur in different areas simultaneously
or sequentially demonstrating that the brain has plasticity (i.e., is malleable). Third, the
8
world is ever changing and older people see events in the context of their unique historical
and cultural lives.
In 1990 Paul and Margaret Baltes proposed a new theory called Selective
Optimization with Compensation Theory (SOC). This theory, now accepted as part of
overall Life Span Theory, hypotheses that the process of “selection, optimization and
compensation” takes place throughout the lives of older adults (Baltes & Baltes, 1990). SOC
theory proposes that three main factors create a successful environment for older adults and
result in successful aging. The theory hypothesizes that a reduction in goals, new goals, or
transformation is a process which helps older adults adjust their expectations in order to take
control of their lives. The second factor, optimization, posits that older adults can continue at
high levels of performance in some areas by employing various strategies (i.e., continuous
practice or new technology). The third factor, compensation, is used when an older adult
employs a method other than their own ability (i.e., a hearing aid or mnemonic strategy) to
overcome a problem. SOC theory includes the notion that adulthood is comprised of many
unique individuals instead of one homogeneous group of individuals. Differences between
these individuals include chronic stress, dementia, frailty, and life expectancy (Baltes &
Smith, 2003). Research has indicated that younger adults are more likely to see their wellbeing in terms of accomplishments and careers, whereas older adults are more likely to link
well-being with good health and the ability to accept change (Baltes & Carstensen, 1996). In
analyses of data from the Berlin Aging Study, Baltes and colleagues reported that SOC
increased over a four-year period particularly among older adults who were rich in resources
versus those with few resources (Freund & Baltes, 2002; Krampe & Baltes, 2003; Lang et al.,
2002).
9
Theories of Cognition
Gerontologists have developed theories suggesting how cognitive decline in
reasoning occurs among older adults. One theory developed by Salthouse (1993)
hypothesizes that speed of processing declines steadily and affects the ability to reason
because it leaves little time for the older adult to complete working memory tasks. Earlier,
Miller (1956) proposed that information was organized into units (e.g., social security
number [534-77-3251] or phone numbers [312-844-1233]) or “chunks,” and that short term
memory is determined by the number of chunks an individual could consciously hold at a
time in working memory (e.g., phone numbers or social security numbers). Chunking is
necessary for holding information in working memory in order to put bits of information
together in various groupings when trying to reason (Naveh-Benjamen et al., 2007). More
recent research have indicated that the ability to hold in mind multiple pieces of information
necessary for reasoning relies on working memory capacity (Morrison, 2005) Older adults
can hold up to four chunks at a time, and older adults take significantly more time to process
information than younger adults (Viskontas, Holyoak, & Knowlton, 2005).
A second theory of age-related cognitive decline developed by Craik and Byrd (1982)
focuses on how the ability to maintain attention impacts inductive reasoning. This theory
hypothesizes that older adult’s experience a decline in the resources necessary to pay
attention and focus when reasoning and that this affects their ability to finish the reasoning
process in the allotted time for increasingly complex tasks. In support of this model recent
research indicates that a task requiring maintenance and manipulation in working memory
affects the ability to reason in older adults more than younger adults (Craik & Byrd, 1982;
Old & Naveh-Benjamin, 2008a, 2008b).
10
Hasher and Zacks (1988) proposed a third theory that suggests as people age they
have difficulty inhibiting irrelevant stimuli and this interferes with the ability to reason. This
theory predicts that as problems become more complex due to the addition of irrelevant
information older adults will experience more trouble maintaining sufficient focus to
complete the task (Hasher & Zacks, 1988; May, Hasher, & Kane, 1999). The role of
inhibition in reasoning is consistent with Baddeley’s (1996) theory characterizing the
executive component of working memory as reflecting the capacity to attend selectively to a
stimulus while inhibiting the disruptive effects of other stimuli. Other researchers suggest
that the ability to inhibit irrelevant information may cause difficulty in reasoning when
problem solving (Robin & Holyoak, 1995). Finally, researchers have found that selective
attention to color supports the inhibitory deficit hypothesis that the age-related increase in the
“Stroop effect” results from a decline in the ability of older adults to inhibit stimuli (West &
Alain, 2000).
Cognitive Decline with Age
Three cognitive abilities have been shown in longitudinal research to exhibit
relatively early age-related decline beginning around 65 years of age (Schaie, 1996). These
are working memory, speed of processing, and reasoning. Previous research indicates that
the brain is malleable or has the capacity to build neural networks over a lifetime including
the later years (Recanzone, 2000). Cognitive stimulation is a predictor of enhancement or
maintenance of cognitive abilities. Moreover, sustained engagement in cognitively
stimulating activities has been found to impact neural structure in both older humans and
rodents (Churchill et al., 2002; Krampe et al., 1996). Some cognitive abilities have been
targeted to see if it is possible to delay the impairment (i.e., decline) of the ability to sustain
11
effective cognitive performance. One of those cognitive abilities is inductive reasoning,
which is the focus of the present research.
Reasoning
A definition of reasoning that is frequently used in the field of aging has been offered
by Salthouse (1996) who describes reasoning as the ability to manipulate different cognitive
abilities (i.e., speed of processing, memory, integration of relationships between information)
in various combinations when presented with novel situations. The cognitive ability to
reason is fundamental to maintenance of an individual’s lifestyle and independence since it is
necessary in the decision making process. The two types of reasoning are deductive and
inductive. Deductive reasoning occurs by drawing inferences or reaching conclusions by
using formal logic. For example: Premise 1: All humans are mortal. Premise 2: Socrates is a
human. Conclusion: Socrates is mortal. Inductive reasoning occurs by making inferences
that are based on previous observations. This type of reasoning can result in inaccurate
predictions or explanations because the accuracy of the result is based on the truth of the
original premise. A common example of this is taken from David Hume, one of the great
philosophers of the 19th century. In his written work, “Enquiry Concerning Human
Understanding,” he postulates the following premise: The sun has risen in the east every
morning until now. Conclusion: The sun will also rise in the east tomorrow (Hume, 1748, p.
4.2). This type of reasoning is used daily but changes over time. Although it is known that
reasoning ability declines with age, the underlying mechanisms are not well understood
(Salthouse, 1992, 2005). As discussed below several theories have been developed by
gerontologists related to the reasoning processes of older adults.
12
Consequences of Declines in Reasoning
The ability to reason is instrumental in the success or failure of an older adult’s ability
to age successfully and maintain his or her independence. Reasoning is used to solve
everyday problems and allows older adults to function in the community and avoid the need
for assistance from both formal (i.e., assisted living or long term care) or informal (e.g.,
spouse, other family members or friends) sources. Reasoning is involved in everyday
functioning and includes such activities as taking medication, balancing a checkbook, or
dressing and grooming. Inductive reasoning allows older adults to gather new information
and create new solutions to problems and thereby avoid loss of independence and the serious
consequences that follow (i.e., accidents, institutionalization, and hospitalizations). In
addition, other losses of specific psychological attributes, such as self-confidence,
motivation, and self-esteem, can impact decision making and reasoning with serious
consequences (e.g., the individual is unable to understand how to turn off a stove burner
resulting in injuries). It is apparent that the loss of the ability to reason can affect all areas of
an older adult’s life (e.g., physical, emotional, and psychological).
Assessment of Reasoning Using Measures of Everyday Functioning
There are several approaches to assessing reasoning in relationship to everyday
cognitive functioning. One approach is to investigate higher order skills (Marsiske &
Margrett, 2006; Marsiske & Willis, 1995). These skills are used by older adults in their
everyday lives and assist them in remaining independent in their homes. These day-to-day
activities include shopping, taking medications, or using transportation (Willis, 1991). This
approach became accepted in the 1980s when a shift in theory led to the view that cognitive
functioning was part of an overall functional health model that consisted of three factors:
13
cognitive, physical and social abilities. Several self-report measures of everyday functioning
have been created to assess activities of day-to-day living such as the Duke Older Americans
Resources and Services (OARS) measure developed by Fillenbaum (1987a, b), the Activity
of Daily Living Scale (ADL) developed by Katz et al. (1963), and the Instrumental Activity
of Daily Living Scale (IADL) developed by Lawton and Brody (1969). In addition,
observational measures have been created to measure older adult’s ability to carry out tasks
in their everyday home setting such as the Observed Tasks of Daily Living (OTDL) scale
(Diehl et al., 1995; Diehl et al, 2005). Some researchers point out that observational
measures allow a more valid picture of an older adult’s everyday functioning in their real
world environment (Marsiske & Margrett, 2006). Both self-report and observational
measures are considered valid and are related to the actual performance of activities and
ratings of functioning by others such as their spouse (Marsiske & Margrett, 2006; Schaie,
1996).
Cognitive problems are a significant predictor of older adults’ difficulties with basic
and instrumental activities of daily living, which can lead to a loss of independence and
increased costs associated with institutionalization both to the individual, their families, and
society as a whole (Burdick et al., 2005). One approach to helping older adults retain their
independence is to implement interventions to delay the onset of decline and/or maintain or
enhance cognitive functioning.
Cognitive Training Interventions (CTIs)
Cognitive training interventions are structured and repeatable ways to teach abilities
(Willis & Schaie, 1986). In general, CTIs have led to improvements in older adult’s
cognitive abilities such as reasoning, spatial orientation, memory, and speed of processing
14
(Ball et al., 2002; Willis et al., 2006). Numerous studies have demonstrated that CTIs can
improve older adults’ performance in the laboratory as well as in selected everyday activities
outside the laboratory (Berch & Wagster, 2004; Hertzog, Kramer, Wilson, & Lindenberger,
2008; Schaie & Willis, 1999). CTIs have successfully shown that older adults can also
benefit from training in collaboration with other older adults (Margrett & Willis, 2006;
Saczynski, Margrett & Willis, 2004a). CTIs have led to a better understanding of how
training can help older adults in their everyday lives (e.g., taking medicine, reading maps of
bus routes, shopping at stores) and have been shown to be specific to the ability on which
older individuals received training (Jobe et al., 2001). Many CTIs use several outcome
measures in order to better demonstrate a training effect than when using only one measure.
This helps increase the validity of the results by demonstrating that individuals who take part
in the training can use these new or enhanced abilities and apply them to new situations
instead of applying training only “to the test” (Schaie, Willis, Hertzog, & Schulenberg, 1987;
Willis & Schaie, 1986).
One important question concerning CTIs involves a better understanding of how
training works to improve reasoning ability. Gains in performance following CTIs have been
theorized to be due to the increased frequency of strategy use. In other words, training is
thought to increase the participant’s ability to apply newly trained strategies to novel
situations by use of transfer (Dunlosky & Hertzog, 1998a). In one theoretical model of
cognitive transfer, Salomon, Perkins and Globerson (1991) hypothesized that transfer
occurred either by several different routes or by a combination of routes which allowed for
variation in how the training is applied to new situations. Salomon and colleagues suggested
that two types of mechanisms are responsible for transference. One is the result of extensive
15
practice which increases stimulus control and the efficiency of learning by the individual.
The second type of transfer, known as mindful abstraction, requires older adults to
consciously apply their reasoning abilities to a deeper level of processing the information
(Carson & Langer, 2006; Salomon et al., 1991). Researchers currently in the field of
reasoning among older adults suggest that mindful abstraction is the mechanism that is
believed to be what occurs during the training process (Dunlosky & Hertzog, 1998b;
Rasmusson, Rebok, Bylsma, & Brandt, 1999). Mindful abstraction can be divided into two
distinct types of transfer: Forward reaching and backward reaching. Forward reaching is
hypothesized to occur when information becomes encoded initially into memory as a general
principle. Backward reaching is thought to happen during the transfer process itself instead
of during the learning process and occurs when the general principle is applied to new
situations. Therefore, it is this transfer process, from training to applying the training to
novel situations, which is thought to help older adults maintain, enhance, or delay the loss of
their reasoning ability.
Cognitive Strategy Training Interventions (CSTIs)
These strategy interventions train individuals to maintain or enhance cognitive
abilities such as memory, speed of processing, and reasoning. There are only a few cognitive
strategy training interventions which are both large in scope and longitudinal in design.
One large ongoing strategy intervention, the Seattle Longitudinal Study (Schaie et al.,
1987), is considered to be one of the most extensive studies of how people develop and
change throughout adulthood. This study collected data on reasoning and other cognitive
abilities among older adults starting in 1956 with 500 adults recruited from the Seattle area
who ranged from 20 to 60 years of age (Schaie, 1999). They subsequently assessed older
16
participants every seven years. As of 2004 a total of 6000 people had participated in the
study. Researchers Willis and Schaie (1986) divided the older adults into two groups based
on their performance on measures of spatial orientation and inductive reasoning (Willis &
Schaie, 1986). One group was the stable group (i.e., scores did not change over time) and the
other the cognitive decline group (i.e., scores showed evidence of significant loss in both
inductive reasoning and spatial orientation). These two groups were then provided training.
Analyses revealed there was improvement in the level of the two abilities for the cognitive
decline group and improved performance in the stable group, resulting in significant training
effects for both spatial orientation and inductive reasoning abilities (Willis & Schaie, 1986).
In 2004 a secondary analysis revealed there was a gain in strategy use specific to
reasoning ability (Saczynski, Willis, & Schaie, 2004b). Training gains were found up to
seven years following the intervention. Analyses revealed that younger participants (i.e., age
65 to 74 years) and those with higher levels of education showed the greatest gains in the use
of strategies. This suggests that increased strategy use by participants in the reasoning
training group may play a role in the gains made by participants on the reasoning measures
(Saczynski, Willis, & Schaie, 2004b, p. 52).
A second major CSTI study called the Advanced Cognitive Training for Independent
and Vital Elderly (ACTIVE) was begun in 1999 with 2800 older adult participants aged 65 to
90 years. This study represents the largest randomized CSTI to date (Jobe et al., 2001). Data
were collected from participants over a 10 year period. Participants in the ACTIVE study
were randomly assigned to one of four groups (i.e., memory, speed of processing, reasoning,
or an untreated control group) and received training on several different cognitive strategies
to see if these interventions would help them maintain, enhance, or delay the onset of decline
17
on the cognitive ability that was the focus of the training. These three cognitive abilities were
chosen on the basis of previous research which suggested these skills were the first to show a
decline with aging (beginning in the mid-60s). In addition, these three abilities are associated
with more complex activities such as balancing a checkbook, doing taxes, or driving a car, all
of which are important for everyday functioning (Schaie, 1996).
The reasoning group was trained on strategies for performing everyday activities (i.e.,
reading a bus schedules, taking medicine and reading food labels). Analyses revealed that
the cognitive training gains were approximately equal to the decline in cognitive abilities that
would have been expected to occur over a 7 to 14 year period of time (Ball et al., 2002). The
reasoning group also had less difficulty performing tasks related to everyday functioning
compared to the control group as measured by the IADL (Ball et al., 2002). This suggests
that reasoning was associated with competency in performing everyday tasks (Ball et al.,
1998, 2002; Diehl et al., 1995, 2005; Willis, 1996).
Results from additional studies examining the effects of the cognitive training
strategies were published more recently. For example, in 2012, a study of memory selfefficacy, inductive reasoning and older adults reported that those with the highest levels of
self-efficacy benefited the most from inductive reasoning training (Payne et al., 2012).
A second study focused on strategy training for the maintenance or enhancement of
memory examining participants with early stage Alzhiemers’ disease (Cherry & SimmonsD'Gerolamo, 2005; Reuter-Lorenz, 2000). Participants were randomly divided into two
groups with one group given an orientation task before the memory strategy training whereas
the second group was not given the orientation task. Participants were trained in a spaced
retrieval technique involving increasing time periods between seeing an object and retrieving
18
the object. Results revealed that participants given the orientation task performed
significantly better than the non-orientation group, suggesting that strategy training may be
beneficial to older adults even with cognitive problems due to early stage dementia.
The third study investigated the long-term effects of mnemonic training on episodic
memory among adults 60 years of age or older with 112 community participants (O’Hara et
al., 2007). Findings revealed that participants who continued use of training strategies
demonstrated long-term improvement in memory suggesting that cognitive training was
beneficial to older adults. However, intervention gains were negatively affected by age of
the participants and duration of the training sessions. That is, those participants who were
older and took longer to learn the strategies did more poorly than younger and faster
participants. Results also showed that those in the training group did better than the control
group and further suggested that strategy training is helpful to the older adult population.
In summary, there have been very few CTSI studies. Most investigations have been
small studies with few participants, used non-healthy older adult participants, and did not
employ a no treatment control group. Of the two large randomized studies, the ACTIVE
intervention was the only one designed to include participants of different races and from
different geographic locations. The findings reported from this study have been extensive.
However, publications to date have examined the relationship between composite scores on
measures of reasoning abilities rather than examining the number of strategies used by
participants before and after training.
Recently data related to the use of reasoning strategies became available from three of
the six training sites included in the ACTIVE study: Indiana University, Wayne State
19
University, and Pennsylvania State University. It is noteworthy that the ACTIVE study
gathered data on strategies that may benefit reasoning ability. These data are rare because
only a handful of randomized strategy studies have been conducted and the participants in the
ACTIVE study were screened for sensory problems, cognitive problems, and disease before
they were randomized into groups to reduce the possibility of declines in reasoning ability
over time due to these problems.
The frequency of reasoning strategies used by participants may be related to proximal
(i.e., enhancement of reasoning ability) and distal (i.e., everyday functioning) outcomes of
the training program. It is important to identify which groups of participants based on such
factors as age, sex, race, or years of education benefit the most from the reasoning strategy
training. It is also important to examine whether or not gains in the use of reasoning
strategies may transfer to everyday functioning.
Research Questions
To address these issues two research questions were examined in this research. First,
did participation in the reasoning training program lead to increased use of strategies taught
to participants? Are there demographic characteristics of participants (i.e., age, sex, race, or
years of education) that moderated the effectiveness of the training?
Second, did participation in the reasoning training program influence changes in
everyday functioning over time (i.e., prior to the intervention to 5 years following the
intervention)? That is, did gains in the use of the reasoning strategies predict changes over
time on two measures of everyday functional status: the Observed Tasks of Daily Living, an
observational measure, and the Instrumental Activities of Daily Living, a self-report
measure?
20
CHAPTER 3. METHODS
Study Design
The data set that was used in the present study was created from a randomized
controlled trial (the ACTIVE study) with older adults who were recruited from six states in
the United States: Alabama, Massachusetts, Maryland, Michigan, Indiana, and Pennsylvania
(Jobe et al., 2001). This geographic dispersion allowed for the inclusion of both rural and
urban older adults from multiple regions of the country. Older adults were recruited in 1998
and 1999 from community centers, clinics, senior housing, and other similar sites. A fourgroup study design was used which included three intervention groups and an untreated
control group. The amount of social contact received by the study’s older adults was equal
for the three training groups. By contrast, there was no contact between participants who
were assigned to the control group. The use of a no-contact control group was based on the
findings from other large clinical trials involving older adults indicating that social contact
among participants did not affect the results for studies on the cognitive and functional status
of participants (Willis et al., 1983, 2006). Each of the six sites randomized their own
participants to one of the four groups: control, memory, speed of processing, and reasoning
(Jobe et al., 2001). Six assessments of participants over time were conducted: baseline
(prior to the intervention), post-intervention (after 10 weeks of training), and annually at 1, 2,
3, and 5, 7, and 10 years after the baseline assessment. In the present analyses data from preintervention, post-intervention, and years 1, 2, 3, and 5 were employed.
21
Participants and Procedures
Participant Eligibility
A total of 5,000 adults 65 years of age and older were contacted regarding
participation in the study, with participants required to be living independently and not have
any problems in terms of cognition, sensory status, and disease state based on initial
screening assessments (Ball et al., 2002). Good cognitive status was determined by a score
of 23 or higher on the Mini Mental State Exam (Folstein, Folstein, & McHugh, 1975). Lack
of functional problems was reflected by a score of 3 or higher on the Activities of Daily
Living (ADL) Scale, a multi-dimensional scale taken from the Older Americans Resources
and Services Scale (OARS; Fillembaum & Smyer, 1981). Participants were tested for
sensory status in terms of deficits in vision, hearing, or communication; individuals with
deficits were excluded from participating to insure that randomized participants would not
have sensory difficulties which would interfere with training. Finally, older adults with
serious medical conditions associated with “imminent functional decline or death” (Willis et
al., 2006) and those who would not be available during the study period were excluded.
Eighteen percent of the older adults originally selected to participate were ineligible for the
study based on the selection criteria and 25.3% refused to participate, resulting in a final
sample of 2832 participants. Thirty of these participants were later dropped from the
analyses due to an error during the randomization process at one of the intervention sites.
Differences Between Participants and Nonparticipants
The 2,802 individuals recruited to participate in the study were younger, more likely
to be White, married, male, better educated, and less likely to have either diabetes or heart
disease than individuals who were found to be ineligible to participate (Willis et al., 2006).
22
Thus, this sample was not a representative sample, but rather a sample of relatively healthy
older adults.
Interviewer Training
Interviewers were trained and certified by first attending a 5-day workshop focusing
on the study assessment protocol at the Coordinating Center of the New England Research
Institute. Interviewers first practiced with each other and then with volunteer older adults in
areas where the intervention was to be conducted (Jobe et al., 2001). Interviewers were
required to practice a specific number of assessments in which they used volunteer
participants who were both younger and older adults. Study coordinators observed and
certified the interviewers during their practice training and gave feedback both verbally and
in written form (Jobe et al., 2001). Finally, interviewers were observed in the field and given
feedback to make sure they were following the protocol (Jobe et al., 2001).
Pre-Intervention Assessment
Eligible participants attended two testing sessions where the reasoning measures
(letter series, letter sets, and word series) and the two measures of functioning (i.e., the
OTDL and the IADL) were administered along with other measures. Completion time for
the first session by the participants ranged from 90 to 120 minutes (Ball et al., 2002). In the
second session, older adults were placed into small groups of approximately 4 to 6 adults.
Assessments of intelligence, cognition, and self- report measures of everyday functioning
(IADL) were conducted at that time which took approximately 3 hours to complete.
Reasoning Strategy Training
Strategy training took place in small groups of three to five participants. There were
a total of 10 sessions that ranged from 60 to 75 minutes in duration which were conducted
23
over a 5 to 6 week period. The first five sessions focused on strategy instruction and
individual and group exercises to practice the strategy (i.e., slash marks, tic marks,
underlining and inserting letter). The second five sessions (e.g., 6 through 10) provided
additional practice with no additional strategies being introduced (Ball et al., 2002). Two
booster training sessions, each lasting 60 to 75 minutes, were conducted at 11 months and 33
months following the baseline assessment for a randomly selected subgroup of the
participants in the reasoning strategy training. All sessions were completed at least two years
before the 5-year follow-up assessment (Willis et al., 2006).
Older adults were trained to use four individual reasoning strategies so they could
apply these strategies later in their everyday functioning tasks. Two measures were used to
assess everyday functioning: One self-report measure, the Instrumental Activities of Daily
Living (IADL), and one observational measure, the Observed Tasks of Daily Living
(OTDL). Each of the four unique strategies helped to assist in inductive reasoning. As can be
seen in Appendix B, four reasoning strategies were taught to participants: slash marks, tic
marks, underlining, or inserting a letter to help participants to physically denote a pattern in
information. During training participants used three measures of inductive reasoning: the
Letter Series, the Letter Sets, and the Word Series. These measures provided participants
with information by which they could discover unique patterns and apply the four training
strategies (i.e., slash marks, tic marks, underlining or letter insertion).
Participants were taught each of the four training strategies by first finding unique
patterns in one of the inductive reasoning measures and then specifically using one of the
four strategies to mark the pattern in the next, novel example. Each training strategy
reflected a unique inductive reasoning skill. To begin training, staff first showed participants
24
what a pattern looked like in an example. This demonstration was both verbal and physical in
nature and used knocking and clapping so participants could see and hear the pattern (i.e.,
knock knock, clap clap). After this first demonstration, participants were trained to scan
information and say it out loud themselves. This was followed by participants learning each
individual strategy one at a time over the course of the total training period. That is, after
training and practice of one strategy, another strategy was then introduced and practiced
during a different training session. Training was conducted using a nested design of 4-5 older
adults who were taught reasoning strategies and practiced for the first 5 sessions followed by
practicing the strategies for the next 5 session (i.e., total of 10 sessions).
The order of the strategies that were taught during the training program was as
follows: underlining, slash marks, tic marks, insert a letter. Each individual strategy was
chosen for a specific reason with substantive cognitive theories related to reasoning (i.e.
inhibition, attention, and working memory) directing choice of strategies when the ACTIVE
study was designed. For example, the underlining strategy helped participants to see smaller
patterns within larger sets of information they scanned during training. The slash mark and
tick mark strategy taught participants how to physically denote patterns in scanned
information to reduce the work load in memory. The tic mark strategy also helped
participants to inhibit information that they typically would use as a cognitive schema and
transform it into a visual object. The insert letter strategy served to help inhibit automatic
responses during their search for patterns of information and to insert a letter as a place
holder for a cognitive schema so they could go on to the next part of finding a pattern. This
strategy also served as a reduction in working memory load while participants sought
patterns of information.
25
Attrition
Willis and colleagues (2006) published information on participation rates for the
ACTIVE study over the first five years. The reasoning and control group dropout rates were
similar for both pre-intervention and post-intervention assessments as well as for the Year 1,
2, 3, and 5 annual assessments. The reasons participants did not complete interviews
included withdrawal from the study, mortality, or the completion of only partial assessments.
Sample
The current study employed data from three of the six sites involved in the ACTIVE
study: Indiana University, Wayne State University, and Pennsylvania State University.
There were a total of 601 participants selected from these three sites who were either
assigned to the reasoning strategy training group (N = 304) or the control group (N = 297).
Fourteen of these participants had incomplete data for post-intervention assessment and were
therefore not included in the analyses, resulting in a final sample of 587 participants.
Measures
Demographic Variables
The following demographic data were collected and included in the analyses: Age in
years, race, sex, and years of educational attainment. In addition, the site from which
participants were recruited was used as a control variable.
Measures of Inductive Reasoning
Three measures of inductive reasoning (i.e., Letter Series, Letter Sets, and Word
Series) were used during the study to measure the reasoning strategies used by participants
before and after training. Assessments of total observed reasoning strategy use were
conducted at each of six time points: Pre-intervention, post-intervention, and annually at
26
years 1, 2, 3 and 5 following the pre-intervention assessment. The present analyses
employed data on strategy use pre-intervention and post-intervention.
Letter Series. This measure was developed to assess inductive reasoning (Blieszner,
Willis, & Baltes, 1981) and consists of 20 items. Participants are required to identify patterns
found in a series of letters. They are asked to indicate what the next letter would be in the
series after they had made the initial pattern determination. The number of letters in each
series ranged from 7 to 15. A series of letters, such as “a b c d e f g h,” was shown on the left
side of the questionnaire. Participants were required to look on the right side of the page to
determine which letter would come next in the series, such as “a b c d e f g h.” The example
given here should result in the correct answer of “i”. Coefficient alpha for the total score for
the measure was reported to be .91 by Blieszner and colleagues (1981). The number of
correct items is summed to create a total score, with higher scores reflecting better
performance on the measure. Participants were given 6 minutes to complete the measure.
Letter Sets. This measure was developed to assess cognitive reasoning (Ekstrom,
French, Harman, & Derman, 1976) and consists of 15 items. Each participant views several
sets of letters and identifies the set that did not use the same pattern rule as the others.
Participants view a page with 15 lines of letter sets with each line consisting of five letters.
Participants then mark an “x” through the letter in the set which they believe does not fit the
rule. For example, if a participant saw a set of five letters such as “acegi bdfhj jlmnp tvxz,”
they should have x’d out the letter set “jlmnp.” This letter set, “jlmnp,” does not follow the
rule of “every other letter in each set.” This was a timed test which participants had 7
minutes to complete. Coefficient alpha was reported to range from .74 to .84 for the measure
27
by Ekstrom and colleagues (1976). The number of correct items is summed to create a total
score with higher scores indicating better performance.
Word Series. This measure was also developed to assess inductive reasoning (Gonda
& Schaie, 1985) and consists of 30 items. Participants are asked to determine the pattern in
the list or series of related words placed vertically on a page. Once that determination is
made they are asked to indicate what word would come next in the series. Each series of
words follows a different rule. For example, in a word series of months of the year
participants moved down a column containing a list of words such as, “January February
February March April April ____”. Participants had to then determine from a separate
column of possible answer choices such as “January February March April May June” what
word would come next in the series (the answer should be “May”). Participants are given 6
minutes to complete the measure. The number of correct items is summed to create a total
score with higher scores indicting better performance on the task. Coefficient alpha was
reported as .89 for the total score on the measure by Gonda and Schaie (1985).
Reasoning Strategies
As seen in Appendix B four specific strategies were taught to improve participant’s
skill in the area of inductive reasoning. The “underlining” strategy trained participants to
insert an underline in reasoning problems to indicate repetitions, skips, and replications. The
“slash mark” strategy trained participants to insert a slash mark or a series of slash marks to
indicate repetitions, skips, and replications. The “tick” mark strategy trained participants to
insert a tick mark or series of tick marks to indicate repetitions, skips, and replications.
Finally, the “letter insertion” strategy trained participants to insert the appropriate letter into
the repetitions, skips, and replications.
28
Everyday Functioning
Two measures of everyday functioning were also administered to participants. These
measures were selected because they assess activities older adults perform daily in order to
take care of themselves and remain independent.
InstrumentalActivities of Daily Living. This measure assessed older adults’ higher
order (i.e., more complex) abilities that are used in everyday functioning (Lawton & Brody,
1969). Ten common activities of daily living were assessed based on self-reports by
participants. Examples of the activities are using a checkbook, shopping, and completing
taxes; these are all complex tasks requiring the ability to reason. For each item the individual
was asked to indicate who much difficulty they would have completing the activity. They
were given three choices: none, a little, or a lot of difficulty. Higher scores on this measure
reflected poorer functional status.
Observed Tasks of Daily Living (OTDL). The Observed Tasks of Daily Living
scale is a performance-based, observational measure assessed three domains of everyday
functioning (Diehl, 1995). The domains include medication use, telephone use, food, and
finances. Examples of the activities are reading a label from a prescription container, looking
up a telephone number in a telephone book, reading a food label; these are all complex tasks
requiring the ability to reason. It should be noted that higher scores on this measure reflected
better functional status.
29
CHAPTER 4. RESULTS
Sample Characteristics
Table 1 presents descriptive statistics for the demographic variables that were
employed in the analyses for the total sample and each of the three data collection sites (i.e.,
Indiana University, Wayne State University, and Pennsylvania State University). The
average age of participants was 73.81 years (SD = 5.73); the average age of participants did
not vary significantly across the three sites, F (2, 584) = 2.33. There were 78.9% female and
21.1% male participants with the proportion of men and women also not varying
significantly across the three sites, χ² (2, N = 587) = .89. Overall 35.6% of the research
participants were African American. Analyses indicated that the proportion of minority
participants varied significantly across the three sites, χ² (2, N = 587) = 164.99, p < .001. The
highest proportion of African American participants was at the Wayne State University site
(66%) whereas only 6% of the participants at Pennsylvania State University were African
American. Finally, the average level of education for participants was 13.24 years (SD =
2.59); 45.8% of participants had graduated from high school. Level of education also varied
significantly across the three study sites. Specifically, participants from Pennsylvania State
University were significantly less educated than participants from the other two sites, F (2,
584) = 39.03, p <.001. Wayne State University and the Indiana University’s participants did
not differ significantly from one another on education.
Prediction of Strategy Use
The first set of analyses examined correlations among the variables to be employed in
the regression analyses predicting strategy use. As can be seen in Table 2, age was found to
be negatively correlate with education, whereas being male correlated positively with
30
education. Race was uncorrelated with age, sex or education. Next, the correlations between
the demographic variables and the three dependent variables (i.e., strategy use for the Letter
Series, Letter Sets, and Word Series measures) both pre- and post-intervention were
examined. The pre- and post-intervention Letter Series measure was positively correlated
with race, indicating that blacks used fewer strategies than whites. By contrast, the Letter
Sets measure was positively correlated only at post-intervention with race as was the Word
Series measure; once again, black participants used fewer strategies than white participants.
Finally, correlations among the three dependent variables (both pre- and post-intervention)
were examined. As expected, the pre-intervention and post-intervention scores on these three
variables were found to be significantly correlated, and the correlations among the measures
of strategy use for the three tasks were also found to be significantly correlated.
The next set of analyses examined the relationship between the predictor variables
(i.e., demographic characteristics of participants, study site, and treatment condition) and the
use of strategies following the intervention. A series of hierarchical regression analyses were
conducted with variables entered into the regression equation in the following order: (a) preintervention strategy use (i.e., for each type of task, such as letter series), (b) demographic
characteristics of participants (i.e., age, sex, education, and race) and study site, and (c)
treatment condition (i.e., intervention vs. control). The final step of the regression analyses
tested whether or not the effect of the treatment condition on strategy use was moderated by
the other predictor variables (e.g., demographic characteristics of participants, study site).
Letter Series
The first hierarchical regression analysis examined the effects of treatment condition
on the number of strategies that were used in solving the letter series task following the
31
reasoning strategy training. The results are presented in Table 3. Participant scores on the
number of strategies measure that was taken prior to the intervention was entered into the
regression analysis in the first step. This measure was found to be a statistically significant
predictor, F (1, 585) = 16.21, p < .001, accounting for 2.7% of the variance in the postintervention measure. In Step 2, the demographic variables of age, education, sex, and race
were entered into the equation along with two dichotomous variables reflecting the site (i.e.,
Wayne State University, Indiana University, or Pennsylvania State University) where data
were collected. This set of variables was also found to be significantly related to the postintervention measure of number of strategies that were used by the participant, F (6,579) =
6.89, p < .001, accounting for 6.5% of the variance. As can be seen in Table 3 the
participant’s age, years of education, and race were all significantly related to the number of
strategies that were used.
In Step 3, the variable reflecting treatment condition (i.e., control versus reasoning
strategy training) was entered into the regression equation. There was found to be a
statistically significant effect of treatment condition on the post-intervention measure of
strategy use after controlling for the effects of the other predictor variables, F (1, 578) =
191.86, p < .001, accounting for 22.6% of the variance in the outcome variable. After
adjusting for the effects of the control variables the mean number of strategies used by the
control group was 0.82, whereas the adjusted mean for the reasoning strategy training group
was 21.58.
Variables reflecting interactions between the demographic variables and treatment
condition were entered into the regression equation in Step 4. These interaction terms were
also found to be significantly related to the post-intervention measure of number of strategies
32
that were used by the participant, F (6, 572) = 7.39, p < .001, accounting for 4.9% of the
variance in the dependent variable. The interactions of age, education, and race with
treatment condition were all found to be statistically significant.
A series of simple effects analyses was conducted to examine these significant
interactions. For the dichotomous variable of race the effect of treatment on the outcome
variable was examined for White and Black participants. There was a statistically significant
difference between the control and reasoning strategy training conditions for both racial
groups. However, the effect of treatment condition on the post-intervention strategies
measure was significantly greater for White participants, b = 23.96, t (5770) = 12.99, p <
.001, than for Black participants, b = 14.74, t (577) = 5.91, p < .001.
For the two continuous variables of age and education the simple effects of treatment
condition was evaluated for high (i.e., +1 SD) and low (i.e., –1 SD) levels of the demographic
variables. The results for education indicated that the effect of the reasoning strategy training
was significantly greater for participants who were more highly educated, b = 20.16, t (577)
= 5.19, p < .001, than for participants who were less educated, b = 9.43, t (577) = 2.57, p =
.01, although there was a statistically significant effect of the intervention for both levels of
education. The results for age indicated that the effect of the reasoning strategy training was
significantly greater for participants who were younger, b = 20.48, t (577) = 5.19, p < .001,
than for participants who were older, b = 9.11 t (577) = 2.57, p = .001, although once again
there was a statistically significant effect of the intervention for both levels of age.
Letter Sets
The second hierarchical regression analysis examined the effects of treatment
condition on the number of strategies that were used in solving the letter sets task following
33
the conclusion of the reasoning strategy training. The results are presented in Table 4.
Participant scores on the number of strategies measure that was taken prior to the
intervention was entered into the regression analysis in the first step. The pre-intervention
measure of strategy use was found to be statistically significant, F (1, 585) = 38.94, p < .001,
accounting for 6.2% of the variance in the post-intervention measure. In Step 2, the
demographic variables and study site were entered into the regression equation. This set of
variables was also found to be significantly related to the post-intervention measure of
number of strategies that were used by the participant, F (7, 579) = 13.76, p < .001,
accounting for 8.0% of the variance. As can be seen in Table 4, the participant’s years of
education was significantly related to the number of strategies that were used.
In Step 3, the variable reflecting treatment condition was entered into the regression
equation. There was found to be a statistically significant effect of treatment condition after
controlling for the effects of the other predictors, F (8, 578) = 26.07, p < .001, accounting for
12.2% of the variance in the outcome variable. After adjusting for the effects of the control
variables the mean number of strategies that were used by the control group was 0.78
whereas the adjusted mean for the reasoning strategy training group was 9.11.
Variables reflecting interactions between the demographic variables and treatment
condition were entered into the regression equation in Step 4. These interaction terms were
found to be significantly related to the post-intervention measure of number of strategies
used, F (14, 572) = 19.85, p < .001, accounting for 6.2% of the variance. The interaction of
years of education with the treatment condition was found to be statistically significant.
A simple effects analysis was conducted to further examine this significant
interaction. The effect of the treatment condition was evaluated for high (i.e., +1 SD) and
34
low (i.e., –1 SD) levels of education. The impact of the reasoning strategy training was
significantly greater for participants who were more highly educated, b = 11.49, t (577) =
5.32, p < .001, than for participants who were less educated, b = .924, t (577) = .45, p =.65,
with the effect for the less educated group being non-significant.
Word Series
The next analysis examined the effects of treatment condition on the number of
strategies that were used in solving the word series task following the reasoning strategy
training. The results are presented in Table 5. In Step 1 of the regression, the preintervention measure of strategy use was found to be statistically significant, F (1, 585) =
30.55, p < .001, accounting for 5% of the variance in the post-intervention measure. In Step
2, the demographic variables and study site were found to be significantly related to the postintervention measure of number of strategies that were used by the participant, F (7, 579) =
8.38, p < .001, accounting for 4.2% of the variance. As can be seen in Table 5 the
participant’s education, race and site were all significantly related to the number of strategies
that were used.
In Step 3, the variable reflecting treatment condition was entered into the regression
equation. There was a statistically significant effect of treatment condition on the postintervention measure of strategy use after controlling for the effects of the other predictor
variables, F (8, 578) = 16.60, p < .001, accounting for 9.5% of the variance. After adjusting
for the effects of the control variables the mean number of strategies that were used by the
control group was 0.40 whereas the mean for the reasoning strategy training group was 7.42.
Variables reflecting interactions between the demographic variables, site, and
treatment condition were entered into the regression equation in Step 4. These interaction
35
terms were also found to be significantly related to the post-intervention measure of number
of strategies that were used by the participant, F (14, 572) = 11.79, p < .001, accounting for
3.7% of the variance in the dependent variable. The interactions of race and site with
treatment condition were all found to be statistically significant.
Simple effects analyses were conducted to examine these significant interactions. For
the dichotomous race variable the effect of treatment condition on the outcome variable was
examined for White and Black participants. There was a statistically significant difference
between the control and reasoning strategy training groups for both racial groups. However,
the effect of treatment condition on the post-intervention strategies measure was significantly
greater for White participants, b = 11.77, t (577) = 6.34, p < .001, than for Black participants,
b = 6.01, t (577) = 2.97, p < .01. Next, the effect of treatment condition on the outcome
variables was examined for the three sites. The results of the simple effects analyses
indicated that the effect of the reasoning strategy training was significantly greater for
participants who were from the Indiana University, b = 6.01, t (577) = 2.97, p < .01, than for
participants from Wayne State University, b = 1.31, t (577) = .75, p = .45, or Pennsylvania
State University, b = 1.24, t (577) = .47, p= .64, with the effect of the intervention being nonsignificant for the latter two sites.
Summary
The results of the analyses indicated that the reasoning strategy training was very
effective in enhancing the use of strategies that were included in the training. Characteristics
of participants such as age, education, and race were also predictive of strategy use following
the intervention. These participant characteristics were found to interact with the treatment
condition in predicting use of the strategies following the intervention, indicating that the
36
intervention was more effective for participants who were younger, better educated, and
White.
Prediction of Functioning Over Time
The hypothesized model shown in Figure 3 was tested by conducting two latent
growth curve modeling analyses. These analyses examined whether the reasoning strategy
training predicted functional status across time (i.e., prior to the reasoning strategy training
and then 1, 2, 3, and 5 years following the baseline assessment). There were three groups of
participants: (a) the control group, (b) participants in the reasoning strategy training group
who did not receive booster training, and (c) participants in the reasoning strategy training
group who also received booster training at 11 and 33 months following completion of the
intervention. The two measures of functional status were the Instrumental Activities of Daily
Living (IADL), a self-report measure, and the Observed Tasks of Daily Living (OTDL), an
observational measure.
Instrumental Activities of Daily Living
The first set of growth curve modeling analyses was conducted to evaluate the pattern
of change over time for the self-report Instrumental Activities of Daily Living (IADL)
measure. Average scores over time on this measure are presented in Figure 4. Growth curve
modeling analyses were conducted using the Mplus 7.0 program (Muthén & Muthén, 2004).
The intercept (i.e., participant’s status on the dependent variable at baseline), linear, and
quadratic terms were estimated. An initial unconditional model was tested, where these three
latent variables were included in the model but there were no predictors. Based on the
criteria described by Hu and Bentler (1999), the results indicated that this model provided a
good fit to the data, χ2 (6, N =587) = 26.04, p < .001, RMSEA = .08, CFI = .97. The mean of
37
the intercept term (M = 1.46) was significantly greater than zero, t (587) = 14.31, p < .001;
this value reflects the average score of participants on the IADL measure prior to the
intervention. For the linear term the mean value for the sample was –.10; this indicates that
scores on the IADL measure declined by .10 points each year following the baseline
assessment. However, this average value did not differ significantly from zero, t (587) = –
.1.11, p = .27. Finally, the average for the quadratic term was .03 which also did not differ
significantly from zero, t (587) = 1.54, p = .12.
These results indicate that we cannot reject the hypothesis that there was no
significant change on the IADL measure over this 5-year period for the sample as a whole.
However, it is also important to evaluate whether or not there are significant individual
differences in the pattern of change over time. An analysis of the variability of the intercept
(s2 = 4.04), linear (s2 = 1.70), and quadratic (s2 = .08) terms indicated they were all
statistically significant (p < .001). Therefore, it appears there were significant individual
differences in the pattern of change over time on the IADL measure.
The next set of analyses examined the association among these three indicators of
change on the IADL over time as well as their correlation with each of the predictor variables
(see Table 6). Participant’s scores on the IADL assessment at baseline (i.e., the intercept
term) were not significantly correlated with the linear or quadratic change over the next 5
years. However, as expected there was a strong correlation between the linear and quadratic
terms indicating that participants who showed the highest rate of linear change over time
were also found to have the lowest level of quadratic change over time (i.e., they were less
likely to show evidence of nonlinear decline over time). The only predictor variable that was
found to be significantly correlated with baseline scores on the IADL measure was age, with
38
older participants reporting greater problems in performing the IADL tasks at the initial
assessment prior to the intervention. Participant education was found to be significantly
related to linear change over the subsequent five years on the IADL measure; higher levels of
education were associated with a lower rate of change. Education was also significantly
related with the quadratic term, with more educated participants found to demonstrate greater
evidence of nonlinear change over time on the IADL measure.
The final set of growth curve analyses on the IADL measure tested the hypothesized
causal model. This model provided a very good fit to the data, χ2 (22, N =587) = 47.09, p =
.001, RMSEA = .04, CFI = .97. The results of these analyses are presented in Tables 7
through 9 and in Figure 5. The predictor variables were found to be significantly related to
the intercept term, t (587) = 2.51, p = .01, accounting for 6.6% of the variance in the baseline
scores. The IADL intercept was significantly predicted by the participant’s age, t (587) =
4.15, p < .001, and whether or not the participant was from the Wayne State site, t (587) = –
2.08, p = .04. Older participants received higher scores on the IADL whereas participants
from the Wayne State received lower scores relative to participants from the other two sites.
The relationship between these predictor variables and linear change on the IADL measure
over time was also statistically significant, t (587) = 1.98, p = .05, accounting for 6.9% of the
variance. The linear term for the IADL measure was significantly related to participant race,
t (587) = –2.01, p = .04, indicating that black participants declined in functioning over time
more quickly than did white participants. Being recruited from the Wayne State site was also
significantly related to linear change on the IADL measure over time, t (587) = –2.71, p =
.01; the decrease in functioning over time was greater for participants from Indiana
University and Pennsylvania State University than for participants from Wayne State
39
University. Finally, the relationship between the predictor variables and the quadratic term
from the growth curve analysis was non-significant, t (587) = 1.66.
Observed Tasks of Daily Living
The second set of growth curve modeling analyses was conducted to evaluate the
pattern of change over time for the Observed Tasks of Daily Living (OTDL) measure.
Average scores over time on this measure are presented in Figure 6. An initial unconditional
model was tested, where three terms (i.e., intercept, linear, and quadratic) were included but
there were no predictors of these latent variables included in the model. The results indicated
that this model provided a good fit to the data, χ2 (6) = 38.19, p < .001, RMSEA = .10, CFI =
.97. The mean of the intercept term (M = 17.28) was significantly greater than zero, t (587) =
95.75, p < .001; this value reflects the average score of participants on the OTDL measure at
baseline prior to the intervention. For the linear term the average value for the sample was –
.74, which was significantly different from zero, t (587) = –3.88, p < .001. This result
indicates that scores on the OTDL measure decreased by an average of .74 points each year
for the sample following the baseline assessment indicating that the level of functioning
decreased over time for the sample as a whole. Finally, the average for the quadratic term
was -.07 which also differed significantly from zero, t (587) = –2.01, p = .04. This latter
result indicates there was some evidence of a curvilinear change on the OTDL measure over
time. As can be seen in Figure 6, scores on the OTDL measure increased from the baseline
to the Year 2 assessment, indicating improvements in functioning. However, scores
decreased between the Year 2 and Year 3 assessment, reflecting the non-linear change in
functional status over time.
40
An analysis of the variability of the intercept (S2 = 14.46), linear (S2 = 10.87), and
quadratic (S2 = .23) terms indicated that they were all statistically significant (p <.001).
Therefore, it appears there are significant individual differences in the pattern of change over
time on the OTDL measure for study participants.
The next set of analyses examined the association among these three indicators of
change on the OTDL over time as well as their correlation with the predictor variables. The
results are presented in Table 10. Participant’s scores on the OTDL assessment at baseline
(i.e., the intercept term) were not significantly correlated with the linear or quadratic change
over the next five years. However, as expected there was a strong correlation between the
linear and quadratic terms indicating that participants who showed the highest rate of linear
change over time were also found to have the lowest level of quadratic change over time (i.e.,
they were less likely so show evidence of nonlinear change) . Participant age, education,
race, and receiving the reasoning strategy training along with the booster sessions were
significantly correlated with scores on the OTDL measure as baseline (i.e., the intercept
term). Older participants received lower scores on the measure at baseline; as expected this
indicates poorer functional status for this group of participants. By contrast, being White and
better educated were associated with better functional status on the OTDL. Finally,
participants who were assigned to receive the intervention with the booster sessions had
higher scores on the OTDL at baseline prior to intervention.
Participant age was the only predictor variable that was found to be significantly
related to linear change over the subsequent five years on the OTDL. Older participants
demonstrated a slower rate of change on the OTDL measure. Finally, none of the predictor
variables were significantly correlated to the quadratic term.
41
Growth curve analyses for the OTDL measure tested the hypothesized causal model.
This model provided a very good fit to the data, χ2 (22) = 62.79, p < .001, RMSEA = .06, CFI
= .97. The results for the model are presented in Tables 11 through 13 and Figure 7. The
predictor variables were found to be significantly related to the intercept term, t (587) = 6.36,
p < .001, accounting for 36% of the variance in the baseline scores. The OTDL intercept was
significantly predicted by participant’s age, t (587) = -7.53, p < .001; race, t (587) = 4.97, p <
.001; and education, t (587) = 8.81, p < .001. Older participants performed more poorly on
the tasks included in the OTDL measure, whereas White and better educated participants
received higher scores. The relationship between these predictor variables and linear or
quadratic change on the OTDL measure over time were non-significant.
Summary
The results of the growth curve modeling analyses indicated there was no evidence of
significant change over time for the sample as a whole on the IADL measure. However,
there was evidence of a significant change over time for the OTDL measure, with the results
indicating that there was evidence of an improvement in functional status from the baseline
to the Year 2 assessment, followed by a decline in functioning over the next three years.
There was also evidence of significant variability across individuals in the change in
functioning over time. The results indicated that age, education, and race were significantly
related to the level of functioning at the pre-intervention assessment and change in
functioning over time. There was no evidence that the reasoning strategy training had an
effect on the change in the functional status of these individuals over time. Therefore, it does
not appear that the increase in use of the reasoning strategies by participants who received
the intervention had an effect on their daily functioning.
42
CHAPTER 5. DISCUSSION AND CONCLUSIONS
The present study analyzed data from participants who were trained in inductive
reasoning strategies in the ACTIVE study, the largest randomized, geographically dispersed,
longitudinal study of older adults at the time it was begun in 1999. This investigation
randomly assigned participants to one of three intervention conditions that focused on
different cognitive abilities (i.e., memory, speed of processing, or reasoning) or to a no
treatment control condition.
Analyses reported here employed data from participants assigned to the reasoning
strategy training or the control condition. Using these data two issues were addressed. First,
what was the effect of the reasoning strategy training on the use of strategies? Second, did
the reasoning strategy training have an impact on the functional status of participants over the
course of five years?
Impact of the Reasoning Strategy Training on Strategy Use
The results of the analyses indicated that the reasoning strategy training was very
effective in enhancing the use of strategies that were included in the training. Although the
results varied somewhat across the three reasoning tasks employed in the study (i.e., letter
series, letter sets, and word series), participants in the reasoning strategy training condition
were significantly more likely to use the strategies than individuals in the control condition.
Specifically, from 31% to 62% of individuals who received the training employed the
strategies in performing the tasks following the intervention whereas only 2% to 4% of the
control participants used these strategies.
Characteristics of participants such as age, education, and race were also predictive of
strategy use following the intervention. As expected, older age was associated with lower
43
strategy use. Similarly, participants with greater education were more likely to employ these
strategies. Finally, there were racial differences found on the use of strategies for two of the
tasks (i.e., letter series and word series), indicating that Black participants were less likely to
use these reasoning strategies than White participants.
Effect of the Reasoning Strategy Training on Functional Status
Two measures of participant functional status (i.e., the IADL and OTDL) were
administered to participants prior to receiving the intervention and then one, two, three, and
five years later. Growth curve modeling analyses were conducted to evaluate the ability of
the reasoning strategies and demographic characteristics of participants to predict change
over time on these measures. For these analyses the reasoning training group was divided
into two groups as some members of the intervention group (N = 167) were randomly
assigned to receive booster training at 11 and 33 months following the pre-intervention
assessment whereas the remaining members of the intervention group (n = 122) did not.
There was no evidence of changes in functional status on the IADL measure for the sample
as a whole, whereas there appeared to be an improvement on the OTDL measure from
baseline to Year 2 followed by a decline in functioning by Year 3. In previous findings from
the ACTIVE study published in 2006, Willis and colleagues found that the reasoning group
reported significantly less difficulty (effect size, 0.29; 99% confidence interval [CI], 0.030.55) in the instrumental activities of daily living (IADL) than the control group, and no
booster effects for everyday functioning. In the 2006 findings, the reasoning group
maintained effects on its specific targeted cognitive ability (i.e., effect size, 0.26 [99% CI,
0.17-0.35]) through 5 years (Willis et al, 2006).
44
For both measures of functional status there was significant variability across
individuals in change over time, indicating that some participants increased in functional
status, some individuals decreased in functional status, and some participants did not change
in functional status over time. The results indicated that age, education, and race were
significantly related to both initial level and change in functioning over time. As expected,
participants who were older, less educated, and Black were lower in functional status at the
pre-intervention assessment and declined more quickly in functioning over time. However,
in contrast to predictions there was no evidence that the reasoning strategy training was
related to change over time on the measures of everyday functioning.
Implications for Reasoning Strategy training
Although only 2% to 6% of participants used the reasoning strategies at the preintervention assessment, training led to 31% to 62% of participants who received the
intervention using the strategies at the post-intervention assessment. These results suggest
that, as predicted by SOC theory, older adults can optimize their performance through
continuous practice (i.e., optimization) of strategies and thereby enhance their reasoning
abilities. In addition, the results suggest that the participants were able to compensate for any
deficits they had in their cognitive abilities. However, this enhancement in the use of these
reasoning strategies did not lead to a significant change in their functional status after
training as measured by the IADL and the OTDL. SOC theory proposes that the individual
trajectories of participants who received the training should vary since older adults represent
a heterogeneous group of individuals due to their life course experiences, genetics, and health
(Baltes & Smith, 2003).
45
In addition, older adults are known to strive for control in their lives and making
choices regarding what they wish to optimize and compensate for is basic to SOC theory by
choosing strategies in the face of illness, disease, and normal declines in everyday
functioning (Heckenhouse, 2011; Rozario, Kidahashi, & DeRienzis, 2011). It should also be
noted that the use of the individual strategies may have affected aspects of cognitive
functioning beyond reasoning abilities. First, it may be that the strategies allowed the older
adults to find patterns in information helping them to chunk information more swiftly than
before training. This may have reduced the load on working memory and freeing up extra
time for processing and manipulation of information during the reasoning process (Morrison,
2005; Salthouse, 1996). The strategy training may have also allowed participants to learn
how to both focus their attention and inhibit other irrelevant information (Hasher & Zacks,
1988; Persad et al., 2002) which may also have enhanced performance of the reasoning tasks.
Finally, it is possible that some participants were able to internalize their training as
they practiced and may therefore have appeared to not be using the reasoning strategies
following training. Such individuals may no longer need to find answers to the reasoning
problems by consciously using marks to answer their questions, but rather used “mindful
abstraction” reflecting a deeper level of reasoning ability (Salomon et al., 1991). They may
have become very good at doing these tasks and no longer needed to mark down on paper the
patterns they found when performing the reasoning tasks. If true the results of reasoning
strategy training may have been greater than found here based on the observed strategy data.
Education
The analyses revealed that those with higher education made greater gains in observed
strategy use. There may be both biological and environmental reasons for this finding. Prior
46
research with older adults into the biological relationship between education and the brain
has demonstrated that different areas of the brain are activated by different levels of
education (Springer et al., 2005; Stern, 2002). Prior research into the environmental
relationship between education and aging has demonstrated that there is a large difference
between the years of education reported by older adults and the quality of that education,
especially among minorities (Elo, 1997; Kirsch et al., 1993; Manly, 2006; National Research
Council on Race and Health Mortality, 2004). In addition, other studies have found that
Black older adults in the United States have reading skills far below their self-reported levels
of education (Albert & Teresi, 1999; Baker et al., 1996; Kave et al., 2012; Manly et al.,
2002). It may be that the levels of education levels reported by participants in this study did
not adequately reflect the same educational abilities across participants.
Cognitive Reserve
Research studies of healthy older adults have previously established that the volume
of white matter in the brain is negatively associated with cognitive tests including those
involving executive functioning (Brickman & Buchsbaum, 2008; Gunning-Dixon et al.,
2008). In a recent study researchers investigated the relationship between cognitive
functioning, brain reserve, and cognitive reserve. Brain reserve is believed to compensate for
pathological change before symptoms are detected and allow individuals to function better in
daily life than those without this reserve. Analyses revealed that brain reserve and lower
volumes of white brain matter were associated with worse cognitive functioning whereas
greater brain reserve and higher volumes of white matter in the brain were associated with
better cognitive functioning and brain reserve (Brickman et al., 2011). When cognitive
functioning was controlled for, individuals with higher brain reserve had significantly greater
47
white matter volume. This suggests that both brain reserve and cognitive reserve appear to
help individuals with their cognitive functioning and to undermine illness and disease in the
brain.
Screening Protocols for Recruitment of Participants
Although strict protocols in screening participants were followed at all locations the
criteria that were employed may have been insufficient to eliminate individuals who were
experiencing mild cognitive impairment. The Mini-Mental State Exam (MMSE) was used
for screening purposes in the ACTIVE study. This measure is actually a mental status
measure and not a measure meant for the purpose of identifying cognitive impairment
(Wadley et al., 2007). One reason individuals experiencing cognitive impairment are tested
by a battery of measures over the course of several days is that multiple assessments help to
better identify persons experiencing even a mild level of cognitive impairment. Although the
MMSE is commonly used in screening for cognitive impairments in community settings, it
may not have been adequate for the recruitment of cognitively intact participants in the
ACTIVE study since it was given only at one point in time. Given that the literature shows
that higher order tasks are lost before lower order tasks (Ashford et al., 1986; Reisberg,
Ferris, de Leon, & Crook, 1982), it may have been more prudent to administer another
measure that evaluated performance on higher order cognitive tasks during the screening and
recruitment process.
Influence of Reasoning Strategies on Everyday Functioning
Overall, these analyses of the reasoning and control groups from the ACTIVE study
demonstrated that this cognitive intervention helped generally healthy older adults to
improve their ability to apply reasoning strategies successfully after approximately 5 to 6
48
weeks of training. Like the larger study sample of 2,802 participants, the current study
results did not demonstrate that the participants were able to apply improved strategy use to
enhance their everyday functioning as measured by both a self-report measure (IADL) and an
observation measure (OTDL) over the first five years of assessments.
It is interesting to note that the self-report measure showed that the older adults
perceived themselves to be maintaining their everyday functioning over time on more
complex tasks of daily living (IADL) rather than showing the expected decline in functioning
with age. However, the perceptions of trained staff in evaluating the participants’
performance on the everyday functioning measure (OTDL) indicated there were
improvements in functional status during the first two years following the baseline
assessment followed by a steep drop in performance between the 2nd and 3rd annual
assessments. The self-reports of the participants appear to be biased toward reporting no
change in their everyday functional status over time. It may be that despite the gains in
strategy use, participants were not applying these new strategies in their daily lives and so no
discernible improvements over time were observed as a function of the training intervention.
It should also be noted that these analyses employed data from the first 5 years of the
10 year ACTIVE study. More time may be necessary to show the effects of the intervention
on the everyday functioning of participants. As previously mentioned, participants were
recruited into the study based on meeting criteria that ensured a high probability of good
health (i.e., good vision and hearing, MMSE score of 23 or higher, living independently).
This screening may have ensured that the participants also had protective characteristics (i.e.,
better cognitive and brain reserve) that lessened the likelihood of declines in functioning over
time. As a consequence, these participants may not have needed to employ these new
49
strategies in their everyday functioning following the intervention. However, this does not
negate the possibility that the strategy training helped participants in their daily lives. Prior
research into problem solving (i.e., reasoning) with participants suffering from clinical levels
of depression has revealed that improved problem solving skills are associated with the
remediation of depression (D’Zurilla et al., 1998). Other important qualitative factors in the
everyday lives of older adults, such as life satisfaction and well-being, may be dependent
upon reasoning abilities and problem solving. That is, the fact that the quantitative analyses
in this study does not show an association between training gains and everyday functioning
does not mean there were not benefits in the lives of the participants who received the
training. Perhaps questions of a qualitative nature would be beneficial before and after
training to determine if there were changes in everyday function on the IADL and the OTDL.
Limitations
One limitation of this study was that we did not examine improvements in
performance on the reasoning measures and how those improvements may have been related
to strategy use. Were there improvements in performance if participants increased their use
of these strategies? Since not all participants who received the intervention also increased
their use of the strategies, it is important to also examine this relationship among those who
received the training. Related to this point, it is unclear if the increased use of these
strategies is related to performance of daily tasks (i.e., shopping, handling finances, driving)
that require the use of reasoning ability. As it stands, we know nothing about the external
validity of this intervention.
Another limitation and future direction is investigating the characteristics of those
participants who did benefit most from the intervention. That is, an investigation into initial
50
functional level, strategy use at baseline, and other characteristics of participants may help to
elucidate how they differ from those who did not benefit from the strategy training.
Although several characteristics of participants affected the amount of observed strategy use
after training, other factors may also moderate these gains such as specific health conditions
or initial pre-intervention scores on the MMSE. In addition, initial everyday functioning
status may also be predictive of the amount of observed strategy gain.
Not all participants benefited from the training to the same degree as other
participants. That is, those who were White, better educated, and in the younger-old (i.e., age
65-74) age group benefited more than Blacks, less educated, and the oldest-old participants.
It may be that having similar characteristics between training staff and participants may be
one way that training to enhance the effectiveness of the training for some of these groups of
individuals.
Future Research
Future research should examine which of the four reasoning strategies (i.e., slash
marks, tic marks, underlining and letter insertion) were used the most by individuals who
demonstrated the greatest gains in performance on the reasoning tasks. It is possible that
some of the four strategies may be used more often by participants and have a greater impact
on gains in reasoning and daily functioning. This information could be used to target future
efforts and help to focus training on specific groups of individuals in the population.
One possibility is that older adults may use a hierarchy of strategies as they go about
trying to solve problems with their inductive reasoning abilities. The strategies which
participants employed in the ACTIVE study may have been used in a specific order or in
different sequences instead of individuals choosing only one to apply to a problem.
51
Although different sequences are important to investigate, combinations of strategies may be
more powerful than using one strategy in a particular sequence at a time. Therefore, it would
be important in future investigations to evaluate if different combinations of strategies were
employed by older adults. Doing this could enhance training sessions for older adults by
making them more efficacious. In addition, certain groups of participants (i.e., younger age
versus older age or less educated versus more educated) may have found one specific
strategy to be most helpful in comparisons to other strategies. As Selection, Optimization,
and Compensation (SOC) theory suggests, older adults may choose particular areas to
optimize and compensate for their declines in functioning based on their own unique history
and environment. This suggests that programs targeting the enhancement of reasoning
ability could benefit from a better understanding of strategy use and choosing those strategies
that older adults prefer in order to age successfully.
Conclusions
Thirty years of research on cognition among the elderly has led to the development of
cognitive strategy interventions in an attempt to maintain or enhance cognitive abilities.
Delaying cognitive decline and impairment is now better understood because of such
interventions. The study of reasoning strategies has important implications for the everyday
functioning of older adults. Results from the current analyses suggest that reasoning training
can create dramatic improvements in the use of reasoning strategies by older adults
regardless of age, race, sex, and education. The extent to which participants may benefit
from training appears to be moderated by these individual characteristics, indicating that the
amount of gain in the use of reasoning strategies depended on these characteristics of
participants. Although transfer of training gains to everyday function does not appear to have
52
occurred during the first five years of the ACTIVE study, it may require a longer period of
time to observe the impact of training gains on everyday functioning in this population. The
results have been supportive of SOC theory by demonstrating that older adults can both
optimize and compensate for a decline in cognitive abilities through strategy training. Older
adults want to age successfully and make adaptations which continue to make their lives
better not just at the immediate moment but throughout their lives.
53
FIGURES
54
55
PreIntervention
Assessment
Letter Series
Reasoning
Measure
PostIntervention
Assessment
Letter Sets
Reasoning
Measure
Word Series
Reasoning
Measure
Everyday
Functioning
IADL OTDL
Training Strategies
Slash Marks
Tick Marks
Underlining
Insert Letter
Figure 3. Impact of observed strategy gains due to reasoning training on everyday
functioning
56
2.5
2
1.5
1
0.5
0
Baseline Year 1
Year 2
Year 3
Figure 4. Instrumental Activities of Daily Living over time
Year 4
Year 5
57
Figure 5. Causal model for the Instrumental Activities of Daily Living (IADL).
58
19.5
19
18.5
18
17.5
17
16.5
16
Baseline Year 1 Year 2 Year 3 Year 4 Year 5
Figure 6. Observed Tasks of Daily Living measured over time
59
Figure 7. Causal model for the Observed Tasks of Daily Living (OTDL).
60
TABLES
Table 1
Demographic Characteristics of ACTIVE Participants by Intervention Site
Variables
Agea
Total sample
(N = 587)
73.81 (5.73)
Indiana
University
(n = 187)
73.70 (5.73)
Wayne State Pennsylvania
University State University
(n = 181)
(n = 219)
73.21 (5.26)
74.39 (5.54)
Womenb
Menb
463 (78.9%)
124 (21.1%)
147 (78.6%)
40 (21.4%)
139 (76.8%)
42 (23.2%)
177 (80.8%)
42 (19.2%)
χ2(5) = 162.79***
Race
White
Black
380 (64.7%)
207 (35.3%)
Education (years) a
13.24 (2.59)
Mean (standard deviation).
Frequency (standard deviation).
***p < .001.
b
F(2, 584) = 2.33
χ2(5) = .98
Sex
a
Differences across
sites
116 (61.0%)
73 (39%)
13.86 (2.80)
60 (33.1%)
121 (66.9)
13.99 (2.75)
206 (94.1)
13 (5.9%)
12.09 (1.73)
F(2,584) = 39.03***
Table 2
Correlations Between Variables
Variablesa
1
2
1. Age
—
2. Sex
.005
—
3. Race
.046
.067
4. Education
–.127**
5. Letter series, pre
–.074
6. Letter series, post
--
.136**
3
4
5
6
7
8
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
--
—
.041
—
.116**
—
–.143**
.030
.125**
.269**
.250**
7. Letter set, pre
–.013
.034
.050
.092*
.390**
.239**
8. Letter set, post
–.078
.028
.130**
.136**
.170**
.509**
.223**
9. Word series, pre
–.040
–.047
.057
.131**
.498**
.316**
.527**
.222**
10. Word series, post
–.185**
.123**
.178**
.204**
.613**
.139**
.681**
—
—
—
Sex was as coded 0 =female, 1 = male; race was coded as 0 = Black, 1 = White; age and education are centered variables;
pre = pre-intervention; post = post-intervention.
*p < .05. ** p < .01.
—
.164**
61
--
.099*
a
10
--
–.038
.014
9
-—
62
Table 3
Hierarchical Linear Regression Results for Letter Series Strategy Use
Step Variable
B
SE
β
t
1
Pre-Intervention Strategy Use
1.26
0.31
.16
4.03***
2
Pre-Intervention Strategy Use
Age at pre-intervention
1.01
–3.66
0.31
0.87
.13
–.17
3.27***
–4.20***
Sex
Race
Years of education
Intervention site: WSU
Intervention site: PSU
–.26
5.39
2.88
–1.08
–1.03
2.13
2.20
0.97
2.26
2.34
–.01
.12
.13
–.02
–.02
–0.12
2.45*
2.97***
–0.48
–0.44
3
4
Pre-Intervention Strategy Use
1.14
0.26
.15
4.33***
Age at baseline
Sex
Race
Years of education
Intervention site: WSU.
Intervention site: PSU
–2.98
1.54
4.48
3.01
–1.94
–0.96
0.76
1.89
1.91
0.84
1.96
2.03
–.14
.03
.10
.14
–.04
–.02
–3.93***
0.83
2.35*
3.58
–0.99
–0.48
Reasoning strategy training
20.70
1.50
.48
13.85***
Pre-Intervention Strategy Use
Age at baseline
Sex
Race
Years of education
Intervention site: WSU
Intervention site: PSU
Reasoning strategy training
Age * Intervention
1.18
–0.27
–0.30
0.46
0.55
–1.08
–.90
14.79
–5.69
0.26
0.26
2.43
1.14
2.56
2.66
2.76
3.41
1.47
.15
.15
–.01
.02
.01
–.02
–.02
.34
–.18
4.51***
4.51***
–0.12
0.41
0.21
–0.41
–0.33
4.34***
–3.86
Sex * Intervention
Race * Intervention
Education * Intervention
WSU * Intervention
4.61
8.55
5.36
–1.66
3.61
3.70
1.63
3.80
.06
.19
.17
–.03
1.28
2.31*
3.30***
–0.44
PSU * Intervention
–0.36
3.94
–.01
–0.09
Note. Sex was coded as 0 = female, 1 = male; race was coded as 0 = Black, 1 = White; intervention site was
coded as 1 = Wayne State University (WSU) and 2 = Pennsylvania State University (PSU), reasoning strategy
training was coded as 0 = control group and 1 = reasoning group; age and education are centered variables; the
sites of WSU and PSU are compared to Indiana University.
*p < .05. ***p < .001.
63
Table 4
Hierarchical Linear Regression Results for Letter Sets Strategy Use
Step Variable
B
SE
β
t
1
Pre-Intervention Strategy Use
0.75
0.12
.25
6.24***
2
Pre-Intervention Strategy Use
Age at pre-intervention
Sex
0.61
–1.23
0.05
0.12
0.46
1.12
.20
–.11
.00
5.16***
–2.60*
0.04
Race
Years of education
Intervention site: WSU.
Intervention site: PSU
2.07
2.73
–1.58
–.63
1.15
0.51
1.18
1.23
.09
.23
–.06
–.03
1.79
5.38***
–1.34
–0.51
0.59
0.11
.20
5.40***
–0.97
0.73
1.75
2.81
–1.95
–0.64
0.43
1.04
1.07
0.47
1.10
1.14
–.08
.03
.07
.24
–.08
–.03
–2.28*
0.70
1.63
5.97***
–1.78
–0.56
8.21
0.84
.35
9.82***
0.78
–0.21
–0.21
0.93
0.25
–1.18
–1.48
6.21
–1.54
0.11
0.56
0.56
1.43
0.64
1.48
1.54
1.90
0.82
.21
–.02
–.02
.04
.02
–.05
–.06
.27
–.09
6.10***
–0.38
–0.38
0.65
0.39
–0.80
–0.97
3.28**
–1.88
Sex * Intervention
Race * Intervention
2.45
2.11
2.01
2.06
.06
.09
1.22
1.03
Education * Intervention
PSU * Intervention
Wayne St. * Intervention
5.28
1.32
0.91
2.19
.31
.04
5.82***
0.60
–1.46
2.11
–.05
3
Pre-Intervention Strategy Use
Age at baseline
Sex
Race
Years of education
Intervention site: WSU
Intervention site: PSU
Reasoning strategy training
4
Pre-Intervention Strategy Use
Age at baseline
Sex
Race
Years of education
Intervention site: WSU
Intervention site: PSU
Reasoning strategy training
Age * Intervention
–0.69
Note. Sex was coded as 0 = female, 1 = male; race was coded as 0 = Black, 1 = White; intervention site was
coded as 1 = Wayne State University (WSU) and 2 = Pennsylvania State University (PSU), reasoning strategy
training was coded as 0 = control group and 1 = reasoning group; age and education are centered variables; the
sites of WSU and PSU are compared to Indiana University.
*p < .05. **p < .01. ***p < .001.
64
Table 5
Hierarchical Linear Regression Results for Word Series Strategy Use
Step Variable
B
SE
β
t
1
Pre-Intervention Strategy Use
1.38
0.25
.22
5.53***
2
Pre-Intervention Strategy Use
Age at pre-intervention
1.24
0.25
.20
4.99***
–0.81
0.47
–.07
–1.72
Sex
0.04
1.14
.00
0.03
Race
2.83
1.18
.12
2.40*
Years of education
1.07
0.52
.09
2.07*
Intervention site: WSU
–2.88
1.21
–.12
–2.39*
Intervention site: PSU
–2.45
1.26
–.10
1.30
0.24
.21
–0.57
0.44
–.05
–1.29
Sex
0.64
1.08
.02
0.59
Race
2.51
1.12
.10
2.25*
Years of education
1.13
0.49
.10
2.29*
Intervention site: WSU
–3.18
1.15
–.13
–2.78*
Intervention site: PSU
–2.43
1.19
–.10
–2.04*
7.19
0.88
.31
1.34
–0.08
0.23
.22
0.59
–.01
–0.14
–0.54
-.13
1.44
1.52
–.02
-.01
–0.28
0.09
.31
.68
.03
.47
Intervention site: WSU
–0.84
1.57
–.03
–0.54
Intervention site: PSU
Reasoning strategy training
–0.27
6.01
1.64
2.02
–.01
.26
–0.17
2.97**
Age * Intervention
-.91
-.05
Sex * Intervention
3.23
.85
2.14
-1.04
1.51
Race * Intervention
5.76
2.20
.23
1.75
–4.77
–4.70
0.97
2.34
0.97
1.03
–.16
.10
3
Pre-Intervention Strategy Use
Age at baseline
Reasoning strategy training
4
Pre-Intervention Strategy Use
Age at baseline
Sex
Race
Years of education
Education * Intervention
PSU * Intervention
WSU * Intervention
.08
–11.95
5.51***
8.21***
57.90**
2.63***
–1.81
–2.04*
1.81
Note. Sex was coded as 0 = female, 1 = male; race was coded as 0 = Black, 1 = White; intervention site was
coded as 1 = Wayne State University (WSU) and 2 = Pennsylvania State University (PSU), reasoning strategy
training was coded as 0 = control group and 1 = reasoning group; age and education are centered variables; the
sites of WSU and PSU are compared to Indiana University.
*p < .05. **p < .01. ***p < .001.
Table 6
Correlations Among the Instrumental Activities of Daily Living (IADL) Model Variables
IADL-I
1
1. IADL-L
–.01
—
2. IADL-Q
–.03
2
–.88***
—
3
4
5
6
7
8
3. Age
.22***
.07
–.04
—
4. Sex
.04
.02
–.02
.01
—
5. Race
.01
–.11
.08
.05
.07
6. Education
–.07
–.14*
.15*
–.13**
.14***
.04
—
7. WSU
–.09
–.12
.19
–.07
.03
.03***
.19***
8. PSU
–.01
–.02
–.01
–.04
.03***
–.34***
9. No booster
.02
–.02
.02
–.04
–.05
–.05
–.01
.01
–.04
10. Booster
.03
–.03
–.011
–.04
–.03
.06
.00
.01
.04
10
—
—
–.52***
65
.08*
9
—
—
–.32***
—
Note. IADL-I = IADL-intercept; IADL-L = IADL-linear; IADL-Q = IADL-quadratic; sex was coded as 0 = female, 1 = male; race was coded as 0 = Black, 1 =
White; intervention site was coded as 1 = Wayne State University (WSU) and 2 = Pennsylvania State University (PSU), reasoning strategy training was coded as
0 = control group and 1 = reasoning group; age and education are centered variables; the sites of WSU and PSU are compared to Indiana University.
*p < .05 **p < .01. ***p <.001.
66
Table 7
Growth Curve Modeling Results for the Instrumental Activities of Daily Living Intercept
b
SE
β
Age
0.08
.02
.21
4.15***
Sex
0.26
.25
.05
1.06
Race
–0.07
.25
–.02
–0.27
Education
–0.05
.04
–.06
–1.06
Intervention site: WSU
–0.54
.26
–.13
–2.08*
Intervention site: PSU
–0.39
.27
–.10
–1.44
Intervention: no booster
0.21
.26
.04
0.80
Intervention: booster
0.27
.23
.06
1.14
Predictor
t
Note. Sex was coded as 0 = female, 1 = male; race was coded as 0 = Black, 1 = White; intervention site was
coded as 1 = Wayne State University (WSU) and 2 = Pennsylvania State University (PSU), reasoning strategy
training was coded as 0 = control group and 1 = reasoning group; age and education are centered variables; the
sites of WSU and PSU are compared to Indiana University.
*p < .05. ***p < .001.
67
Table 8
Growth Curve Modeling Results for the Instrumental Activities of Daily Living Linear Term
b
SE
β
t
Age
0.01
.02
.05
.82
Sex
0.14
.21
.05
.69
Race
–0.43
.21
–.16
–2.01*
Education
–0.06
.04
–.13
–1.75
Intervention site: WSU
–0.60
.22
–.21
–2.71**
Intervention site: PSU
–0.27
.23
–.10
–1.17
Intervention: no booster
–0.11
.22
–.04
–.52
Intervention: booster
–0.05
.20
–.02
–.26
Predictor
Note. Sex was coded as 0 = female, 1 = male; race was coded as 0 = Black, 1 = White; intervention site was
coded as 1 = Wayne State University (WSU) and 2 = Pennsylvania State University (PSU), reasoning strategy
training was coded as 0 = control group and 1 = reasoning group; age and education are centered variables; the
sites of WSU and PSU are compared to Indiana University.
*p < .05. **p < .01.
68
Table 9
Growth Curve Modeling Results for the Instrumental Activities of Daily Living Quadratic Term
b
SE
β
t
Age
0.00
.00
–.02
–.30
Sex
–0.04
.05
–.05
–.77
Race
0.07
.05
.12
1.50
Education
0.02
.01
.14
1.99*
Intervention site: WSU
0.10
.05
.15
1.99*
Intervention site: PSU
0.04
.05
.06
.77
Intervention: no booster
0.02
.05
.02
.34
–0.01
.04
–.02
–.27
Predictor
Intervention: booster
Note. Sex was coded as 0 = female, 1 = male; race was coded as 0 = Black, 1 = White; intervention site was
coded as 1 = Wayne State University (WSU) and 2 = Pennsylvania State University (PSU), reasoning strategy
training was coded as 0 = control group and 1 = reasoning group; age and education are centered variables; the
sites of WSU and PSU are compared to Indiana University.
*p < .05.
Table 10
Correlations Among the Observed Tasks of Daily Living (OTDL) Model Variables
IADL-I
1
1. IADL-L
–.04
—
2. IADL-Q
.09
–.94***
—
–.16**
.10
—
3. Age
–.35***
2
3
4
5
6
4. Sex
.01
.03
–.78
.01
—
5. Race
.27***
.09
–.05
.05
.07
6. Education
.43***
–.04
.01
–.13**
.14***
–.07
.03
–.44***
.19***
–.04
.47***
–.34***
7
8
.04
—
–.04
8. PSU
.02
.01
.05
–.05
.07
–.12
–.04
–.05
–.05
–.01
.01
–.04
–.06
.13
–.04
–.03
.06
–.00
.01
.04
10. Booster
.10*
—
–.52***
—
—
–.06***
—
Note. IADL-I = IADL-intercept; IADL-L = IADL-linear; IADL-Q = IADL- quadratic; WSU = Wayne State University; PSU = Pennsylvania State University;
age and education are centered variables.
*p < .05 **p < .01 ***p <.001.
.
69
–.05
1
–.01
9. No booster
10
—
7. WSU
.08*
9
70
Table 11
Growth Curve Modeling Results for the Observed Tasks of Daily Living Intercept
b
SE
β
Age
–0.21
.03
–.31
–7.53***
Sex
–0.57
.38
–.06
–1.49
Race
1.94
.39
.25
4.97***
Education
0.58
.07
.40
8.81*
Intervention site: WSU
0.29
.40
.04
0.72
Intervention site: PSU
0.57
.42
.07
1.36
–0.21
.40
–.02
–0.52
0.51
.36
.06
1.42
Predictor
Intervention: no booster
Intervention: booster
t
Note. Sex was coded as 0 = female, 1 = male; race was coded as 0 = Black, 1 = White; intervention site was
coded as 1 = Wayne State University (WSU) and 2 = Pennsylvania State University (PSU), reasoning strategy
training was coded as 0 = control group and 1 = reasoning group; age and education are centered variables; the
sites of WSU and PSU are compared to Indiana University.
*p < .05.
71
Table 12
Growth Curve Modeling Results for the Observed Tasks of Daily Living Linear Term
Predictor
b
SE
β
t
Age
–0.10
.03
–.17
–2.94**
Sex
0.24
.47
.03
0.52
Race
0.85
.48
.12
1.76
Education
–0.11
.08
–.09
–1.40
Intervention site: WSU
–0.18
.49
–.03
–0.36
Intervention site: PSU
–0.50
.51
–.07
–0.97
0.44
.49
.06
0.90
–0.35
.44
–.05
–0.79
Intervention: no booster
Intervention: booster
Note. Sex was coded as 0 = female, 1 = male; race was coded as 0 = Black, 1 = White; intervention site was
coded as 1 = Wayne State University (WSU) and 2 = Pennsylvania State University (PSU), reasoning strategy
training was coded as 0 = control group and 1 = reasoning group; age and education are centered variables; the
sites of WSU and PSU are compared to Indiana University.
**p < .01.
72
Table 13
Growth Curve Modeling Results for the Observed Tasks of Daily Living Quadratic Term
b
SE
β
t
Age
0.01
.01
.11
1.48
Sex
–0.09
.09
–.08
–1.04
Race
–0.14
.09
–.14
–1.58
0.02
.02
.09
1.07
Intervention site: WSU
–0.07
.09
–.06
–0.72
Intervention site: PSU
0.10
.10
.09
0.99
–0.10
.09
–.08
–1.09
0.12
.08
.11
1.46
Predictor
Education
Intervention: no booster
Intervention: booster
Note. Sex was coded as 0 = female, 1 = male; race was coded as 0 = Black, 1 = White; intervention site was
coded as 1 = Wayne State University (WSU) and 2 = Pennsylvania State University (PSU), reasoning strategy
training was coded as 0 = control group and 1 = reasoning group; age and education are centered variables; the
sites of WSU and PSU are compared to Indiana University.
73
APPENDIX A. INSTITUTIONAL REVIEW BOARD APPROVAL
74
75
APPENDIX B. FOUR REASONING TRAINING STRATEGIES
Strategy
Example of Training Strategy
Underlining
aab ddeg
Slash Marks
ccd/ffg/i
Tick Marks
dde’ggh’j
Inserted Letters
kkl/mnno/pq
76
REFERENCES
Albert, S. M. & Teresi, J.A. (1999). Reading ability, education and cognitive status
assessment among older adults in Harlem, New York City. American Journal of
Public Health, 89, 95-97. doi: 10.2105/AJPH.89.1.95
Ashford, J., Hsu, L., Becker, M., Kuman, V., & Bekian, C. (1986). Mini-mental status and
activities of daily living: Cross validation by scalogram and item analysis techniques.
The Gerontologist, 26, 143A.
Baddeley, A. (1996). Exploring the central executive. Quarterly Journal of Experimental
Psychology: Human Experimental Psychology, 49, 5–28.
doi:10.1080/027249896392784
Ball, K., Berch, D. B., Helmers, K. F., Jobe, J. B., Leveck, M. D., Marsiske, M., . . . Willis,
S. L. (2002). Effects of cognitive training interventions with older adults: A
randomized controlled trial. Journal of the American Medical Association, 288,
2271–2281. doi:10.1001/jama.288.18.2271
Ball, K., Owsley, C., Stalvey, B., Roenker, D. L., Sloane, M. E., & Graves, M. (1998).
Driving avoidance and functional impairment in older drivers. Accident Analysis &
Prevention, 30(3), 313–322. doi: 10.1016/S0001-4547(97)00102-4
Baltes, P. B. (1987). Theoretical propositions of life-span developmental psychology: On the
dynamics between growth and decline. Developmental Psychology, 23, 611–626.
doi:10.1037//0012-1649.23.5.611
Baltes, P. B., & Baltes, M. M. (1990). Psychological perspectives on successful aging: The
model of selective optimization with compensation. In P. B. Baltes & M. M. Baltes
(Eds.), Successful aging: Perspectives from the behavioral sciences (pp. 1–34). New
77
York, NY: Cambridge University Press. ISBN: 052143582X /ISBN-13:
9780521435826
Baltes, P. B., & Baltes, M. M. (1997). A systemic-holistic view of psychological functioning
in very old age: Introduction to a collection of articles from the Berlin aging study.
Psychology and Aging, 12(3), 395–409. doi: 10.1037/0882-7976.12.3.395
Baltes, M. M., & Carstensen, L. L. (1996). The process of successful aging. Aging and
Society, 16, 397–422. doi:10.1017/S0144686X00003603
Baltes, P. B., & Smith, J. (2003). New frontiers in the future of aging: From successful aging
of the young old to the dilemmas of the fourth age. Gerontology, 49, 123–135.
doi:10.1159/000067946
Baker, F. M., Johnson, J. T., Velli, S. A., (1996). Congruence between education and reading
levels of older persons. Psychiatry Services, 47(2), 194-196. PMID: 8825260
Berch, D. B., & Wagster, M. V. (2004). Future directions in cognitive aging research:
Perspectives from the National Institute on Aging, In R. A. Dixon, L. Backman, & L.
Nilson (Eds.), New frontiers in cognitive aging (p. 333–356). New York, NY: Oxford
University Press. doi: 10.1093/acprof:oso/9780198525691.003.0017
Blieszner, R., Willis, S. L., & Baltes, P. B. (1981). Training research in aging on the fluid
ability of inductive reasoning. Journal of Applied Developmental Psychology, 2, 247–
265. doi:10.1016/0193-3973(81)90005-8
Breitner, J. C. S., Wyse, B. W., Anthony, J. C., Welsh-Bohmer, K. A., Steffens, D. C.,
Norton, M. C., . . . Khachaturian, A. (1999). APOE-epsilon4 count predicts age when
prevalence of AD increases, then declines: The Cache County study. Neurology, 53,
321–331. doi: 10.1212/WNL.53.2.321
78
Brickman, A. M., & Buchsbaum, M.S. (2008). Alzheimer's disease and normal aging:
neurostructures. J.H. Byrne (Ed.), Learning and Memory: A Comprehensive
Reference, Elsevier, New York (2008), pp. 601–621.doi: 10.1016/B978-0123705099.00099-1
Brickman, A. M., Siedlecki, K. L., Muraskin, J., Manly, J. J., Luchsinger, J. A., Yeung, L.,
Brown, T. R., DeCarli, C., Stern, Y. (2011). White matter hyper-intensities and
cognition: Testing the reserve hypothesis. Neurobiology of Aging, 23(9), 1588-1598.
doi:10.1016/j.neurobiolaging.2009.10.013
Burdick, D. J., Rosenblatt, A., Samus, Q. M., Steele, C., Baker, A., Harper, M., . . . Lyketsos,
C. G. (2005). Predictors of functional impairment in residents of assisted living
facilities: The Maryland Assisted Living Study. Journal of Gerontology: Biological
Sciences &Medical Sciences, 60A, 258–264. doi:10.1093/gerona/60.2.258
Callahan, C. M., Hendrie, H. C., & Tierney, W. M. (1995). Documentation and evaluation of
cognitive impairment in elderly primary care patients. Annals of Internal Medicine,
122(6), 422–429. doi:10.7326/0003-4819-122-6-199503150-00004
Carson, S. H., & Langer, E. J. (2006). Mindfulness and self-acceptance. Journal of RationalEmotive and Cognitive-Behavior Therapy, 24(1), 29-43. doi: 10.1007/s10942-0060022-5
Centers for Disease Control and Prevention & The Merck Company Foundation. (2007). The
state of aging and health in America 2007. Whitehouse Station, NJ: The Merck
Company Foundation. Retrieved from http://www.cdc.gov/Aging/pdf/saha_2007.pdf
79
Cherry, K. E., & Simmons-D’Gerolamo, S. S. (2005). Long-term effectiveness of spacedretrieval memory training for older adults with probable Alzheimer’s disease.
Experimental Aging Research, 31(3), 261–289. doi: 10.108003610730590948186
Churchill, J. D., Galvez, .R, Colcombe, S., Swain, R .A , Greenough, W. T. (1982).
Exercise, experience, and the aging brain. The Neurobiology of Aging, 23 (5), 941–
951. doi: S0197-4580(02)00028-3
Craik, F. I. M., & Byrd, M. (1982). Aging and cognitive deficits: The role of attentional
resources. In F. I. M. Craik & S. Trehub (Eds.), Aging and cognitive processes (pp.
191–211). New York, NY: Plenum Press. ISBN-13: 9781468441802
Crimmins, E.M., & Beltrán-Sánchez, H. (2010). Mortality and morbidity trends: is there
compression of morbidity? Journal of Gerontology: Social Sciences,66B,(1), 75–86.
doi:10.1093/geronb/gbq088.
DeCarli, C. (2003). Mild cognitive impairment: Prevalence, prognosis, etiology, and
treatment. Lancet Neurology, 2(1), 15–21. doi: 10.1016/S1474-4422(03)00262-X
Diehl, M., Marsiske, M., Horgas, A. L., Rosenberg, A., Saczynski, J. S., & Willis, S. L.
(2005). The observed tasks of daily living: A performance-based assessment of
everyday problem solving in older adults. Journal of Applied Gerontology, 24(3),
211–230. doi: 10.1177/0733464804273772
Diehl, M. K., Willis, S. L., & Schaie, K. W. (1995). Everyday problem solving in older
adults: Observational assessment and cognitive correlates. Psychology & Aging, 10,
478–491. doi:10.1037//0882-7974.10.3.478
Dixon, R. A., Backman, L., & Lars-Goran, N. (2004). Don’t fence us in: Probing the
frontiers of cognitive aging. In R. Dixon & L. Beckman (Eds.), New frontiers in
80
cognitive aging (pp. 3–4). New York, NY: Oxford University Press. ISBN-0 19
852569 9
Dunlosky, J., & Hertzog, C. (1998). Aging and deficits in associative memory: What is the
role of strategy production? Psychology and Aging, 13(4), 597–607. doi:
10.1037/0882-7974.13.4.597
Dunlosky, J., & Hertzog, C. (1998). Training programs to improve learning in later
adulthood: Helping older adults educate themselves. In D. J. Hacker, J. Dunlosky, &
A. C. Graesser (Eds.), Metacognition in educational theory and practice (pp. 249–
276). Hillsdale, NJ: Erlbaum. ISBN-1410602354, 9781410602350
D'Zurilla, T. J., Maydeu-Olivares, A., & Kant, G. L. (1998). Age and gender differences in
social problem-solving ability. Personality and individual differences, 25(2), 241252. doi: 10.1016/A0191-8869(98)00029-4
Ekstrom, R., French, J., Harman, H., & Derman, D. (1976). Kit of factor-referenced cognitive
tests. Princeton, NJ: Educational Testing Service. Retrieved from
http://www.getcited.org/pub/101805316
Ekstrom, R. B., French, J. W., & Harman, H. H. (1979). Cognitive factors: Their
identification and replication. Multivariate Behavioral Research Monographs, 79(2),
3–84. Retrieved from
http://www.eric.ed.gov/ERICWebPortal/search/detailmini.jsp?_nfpb=true&_&ERIC
ExtSearch_SearchValue_0=EJ220217&ERICExtSearch_SearchType_0=no&accno=
EJ220217
Elder, G. H., Jr. (1992). Models of the life course. Contemporary Sociology, 21(5), 632–635.
doi:10.2307/2075543
81
Elo, I.T. & Preston, S.H. (1997). Racial and ethnic differences in mortality at older ages. In:
Martin, L. G. & Soldo, B. J., (eds). Racial and ethnic differences in the health of older
Americans. Committee on Population, Commission on Behavioral and Social
Sciences and Education, National Research Council (p. 10-14). PMID: 8745107
Federal Interagency Forum on Aging Related Statistics. (2007). Older Americans 2007: Key
indicators of well-being. Washington, DC: U.S. Government Printing Office.
Retrieved from http://www.ncbi.nlm.nih.gov//books/NBK25528
Ferri, C. P., Prince, M., Brayne, C., Brodaty, H., Fratiglioni, L., Ganguli, M., . . . Scazufca,
M. (2005). Global prevalence of dementia: a Delphi consensus study. Lancet, 366,
2112–2117. doi:10.1016/S0140-6736(05)67889-0
Fillenbaum, G. G. (1987a). Multidimensional functional assessment. In G. L. Maddox (Ed.),
The encyclopedia of aging (pp. 460–464). New York, NY: Springer. ISBN: 0-80580241-X
Fillenbaum, G. G. (1987b). OARS Multidimensional Functional Assessment Questionnaire.
In G. L. Maddox (Ed.), The encyclopedia of aging (pp.496–497). New York, NY:
Springer. ISBN: 0-8058-0241-X
Fillenbaum, G. G., & Smyer, M. M. (1981). The development, validity, and reliability of the
Oars Multidimensional Functional Assessment Questionnaire. Journal of
Gerontology, 36(4), 428–434. doi:10.1093/geronj/36.4.428.
Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). Mini-mental state: A practical
method for grading the cognitive state of patients for the clinician. Journal of
Psychiatric Research, 12, 189–198. doi:10.1016/0022-3956(75)90026-6
82
Freedman, V. A., Martin, L. G., & Schoeni, R. F. (2002). Recent trends in disability and
functioning among older adults in the United States: A systemic review. Journal of
the American Medical Association, 288(24), 3137–3146.
doi:10.1001/jama.288.24.3137
Freund, A. M., & Baltes, P. B. (2002). Life-management strategies of selection, optimization
and compensation: Measurement by self-report and construct validity. Journal of
Personality and Social Psychology, 82(4), 642–662. doi:10.1037//00223514.82.4.642
Gonda, J., & Schaie, K.W. (1985). Schaie-Thurstone mental abilities test: Word series test.
Palo Alto, CA: Consulting Psychologists Press.
Gunning-Dixon, F. M., Brickman, A. M., Cheng, J. C., Alexopoulos, G. S. (2008). Aging of
cerebral white matter: A review of MRI findings. International Journal of Geriatric
Psychiatry, 24, 109-117. DOI: 10.1002/gps.2087
Hasher, L., & Zacks, R. T. (1988). Working memory, comprehension, and aging: A review
and a new view. In G. H. Bower (Ed.), The psychology of learning and motivation
(pp. 193–225). San Diego, CA: Academic Press. ISBN-13: 9780125433242
Health and Retirement Study. (2012). (grant number NIA U01AG009740). Ann Arbor:
Michigan. Retrieved from http://hrsonline.isr.umich
.edu/modules/meta/adams/desc/ADAMSSampleWeights_Nov2007.pdf
Heckhausen, J. (2011). Agency and control striving across the life span. In K. Fingerman, C.
Berg, T. Antonucci, & J. Smith (Eds.), Handbook of lifespan psychology (pp. 183–
212). New York, NY: Springer. ISBN: 0826110797 ISBN-13: 9780826110794
83
Heeringa, S. G., Fisher, G. G., Hurd, M. D., Langa, K. M., Ofstedal, M. B., Plassman, B. L.,
. . . Weir, D. R.(2007). Aging, demographics and memory study (ADAMS): Sample
design, weights, and analysis for ADAMS. Retrieved from http://hrsonline.isr.umich
.edu/modules/meta/adams/desc/ADAMSSampleWeights_Nov2007.pdf
Hertzog, C., Kramer, A. F., Wilson, R. S., & Lindenberger, U. (2008). Enrichment effects on
adult cognitive development: Can the functional capacity of older adults be preserved
and enhanced? Psychological Science in the Public Interest, 9, 1–65. doi:10.1111/
j.1539-6053.2009.01034.x
Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure
analysis: Conventional criteria versus new alternatives. Structural Equation
Modeling: A Multidisciplinary Journal, 6(1), 1–55. doi:
10.1080/10705519909540118
Hume, D. (1910). An Enquiry into human understanding. Harvard Classics, 37, Enquiry 4.2.
Copyright 1910 P.F. Collier & Son. Retrieved from http://18th.eserver.org/humeenquiry.html#3.
Jobe, J. B., Smith, D. M., Ball, K., Tennstedt, S. L., Marsiske, M., Willis, S. L., ... &
Kleinman, K. (2001). ACTIVE: A cognitive intervention trial to promote
independence in older adults. Controlled clinical trials, 22(4), 453-479. Doi:
10.1016/S0197-2456(01)00139-8
Katz, S., Ford, A. B., Moskowitz, R. W, Jackson, B. A., Jaffee, M. W, & Cleveland, M. A.
(1963). Studies of illness in the aged. The index of ADL: A standardized measure of
biological and psychosocial function. Journal of the American Medical Association,
185, 914–919. doi:10.1001/jama.1963.03060120024016.
84
Kavé, G., Shrira, A., Palgi, Y., Spalter, T., Ben-Ezra, M., & Shmotkin, D. (2012). Formal
education level versus self-rated literacy as predictors of cognitive aging. The
Journals of Gerontology, Series B: Psychological Sciences and Social Sciences,
67(6), 697–704, doi:10.1093/geronb/gbs031.
Kirsch, I., Jungeblut, A., Jenkins, L., Kolstad, A. Adult Literacy in America: A First Look at
the Results of the National Adult Literacy Survey. Washington, DC: National Center
for Education Statistics, US Department of Education; 1993. NCJ: 144980, 065-00000588-3
Krampe, R. T., & Baltes, P. B. (2003). Intelligence as adaptive resource development and
resource allocation: A new look through the lenses of SOC and expertise. In R. J.
Sternberg (Ed.). The psychology of abilities, competencies, and expertise (pp. 31–68).
New York, NY: Cambridge University. ISBN: 0521007763 ISBN-13:
9780521007764
Krampe, R. T., Ericcson, K A. (1996). Maintaining excellence: deliberate practice and elite
performance in young and older pianists. Journal of Experimental Psychology,
125:331-359. doi: 10.1037/0096-3445.125.4.331
Lang, F. R., Rieckmann, N., & Baltes, M. M. (2002). Adapting to aging losses: Do resources
facilitate strategies of selection, compensation, and optimization in everyday
functioning? Journals of Gerontology: Psychological Sciences & Social Sciences,
57(6), 501–509. doi:10.1093/geronb/57.6.P501
Langa, K. M., Plassman, B. L., Wallace, R. B., Herzog, A. R., Heeringa, S. G., Ofstedal, M.
B., . . . Willis, R. J. (2005). The aging, demographics, and memory study: Study
design and methods. Neuroepidemiology, 25, 181–191. doi:10.1159/000087448
85
Lawton, M. P., & Brody, E. M. (1969). Assessment of older people: Self-maintaining and
instrumental activities of daily living. The Gerontologist, 9(3, Pt 1), 179–186.
doi:10.1093/geront/9.3_Part_1.179
Manly, J.J. (2006). Deconstructing race and ethnicity: implication for measurement of health
outcomes. Medical Care, 44(11), S1–0–S16. doi:
10.1097/01.mlr.0000245427.22788.be
Manly, J. J., Jacobs, D. M., Touradi, P. (2002). Reading level attenuates differences in
neuropsychological test performance between African American and white elders.
Journal of International Neuropsychology Society, 8, 341–348. doi:
10.1017/s1355617702813157
Margrett, J. A., & Willis, S. L. (2006). In-home cognitive training with older married
couples: Individual versus collaborative learning. Aging Neuropsychology and
Cognition, 13, 173–195. doi:10.1080/138255890969285
Marsiske, M., & Margrett, J. A. (2006). Everyday problem solving & decision making. In J.
E. Birren & K. W. Schaie (Eds.). Handbook of psychology of aging (pp. 315–342).
New York, NY: McMillian Press. ISBN-13: 9780123808837
Marsiske, M., & Willis, S. L. (1995). Dimensionality of everyday problem solving in older
adults. Psychology and Aging, 10, 269–283. doi:10.1037//0882-7974.10.2.269
May, C. P., Hasher, L., & Kane, M. J. (1999). The role of interference in memory span.
Memory & Cognition, 27, 759–767. doi:10.3758/BF03198529
Miller, G. A. (1956). The magical number seven, plus or minus two: Some limits on our
capacity for processing information. Psychological Review, 63, 81–97.
doi:10.1037/h0043158
86
Morrison, R. G. (2005). Thinking in working memory. In K. J. Holyoak & R. G. Morrison
(Eds.), Cambridge handbook of thinking and reasoning (pp. 457–473). Cambridge,
UK: Cambridge University Press. ISBN-13: 9780521824170
Muthén, L. K., & Muthén, B. (2004). MPlus: The comprehensive modeling program for
applied researchers, Version 3 [Computer software]. Los Angeles, CA: Muthén &
Muthén .
National Institute on Aging. (2012). Minority aging and health disparities. Retrieved from
http://www.nia.nih.gov/about/minority-aging-and-health-disparities
National Research Council (US) Panel on Race, Ethnicity, and Health in Later Life; N. B.
Anderson, R. A. Bulatao, B. Cohen, editors. Critical Perspectives on Racial and
Ethnic Differences in Health in Late Life. Washington (DC): National Academies
Press (US); 2004. Racial and Ethnic Disparities in Health and Mortality Among the
U.S. Elderly Population. Retrieved from:
http://www.ncbi.nlm.nih.gov/books/NBK25528/
Naveh-Benjamen, M., Cowan, N., Kilb, A., & Chen, Z. (2007). Age-related differences in
immediate serial recall: Dissociating chunk formation and capacity. Memory and
Cognition, 35(4), 724–737 doi:10.3758/BF03193310
Old, S. R., & Naveh-Benjamin, M. (2008a). Differential effects of age on item and
associative measures of memory: A meta-analysis. Psychology and Aging, 23(1),
104–108. doi: 10.1037/0882-7974.23.1.104
Old, S. R., & Naveh-Benjamin, M. (2008b). Memory for people and their actions: Further
evidence for an age-related associative deficit. Psychology and Aging, 23(2), 467–
472. doi: 10.1037/0882-7974.23.2.467
87
Payne, B. R., Jackson, J. J., Hill, P. L., Gao, X., Roberts, B. W., & Stine-Morrow, E. A.
(2012). Memory self-efficacy predicts responsiveness to inductive reasoning training
in older adults. The Journals of Gerontology Series B: Psychological Sciences and
Social Sciences, 67(1), 27-35. doi: 10.1093/geronb/gbr073
Persad, C. C., Abeles, N., Zacks, R. T., & Denburg, N. L. (2002). Inhibitory changes after
age 60 and their relationship to measures of attention and memory. Journal of
Gerontology: Psychological Sciences, 57B, 223–232. doi: 10.1093/geronb/57.3.P22
Petersen, R. C., Smith, G. E., Waring, S. C., Ivnik, R. J., Tangalos, E. G., & Kokmen, E.
(1999). Mild cognitive impairment: clinical characterization and outcome. Archives of
Neurology, 56(3), 303–308. doi:10.1001/archneur.56.3.303
Plassman, B. L., Langa, K. M., Fisher, G. G., Heeringa, S. G., Weir, D. R., Ofstedal, M. B.,
. . . Wallace, R. B. (2007). Prevalence of dementia in the United States: The aging,
demographics, and memory study, Neuroepidemiology, 29(1–2), 125–132.
doi:10.1159/ 000109998
Plassman, B. L., Langa, K. M., Fisher, G. G., Heeringa, S. G., Weir, D. R., Ofstedal, M. B.,
. . . Wallace, R. B. (2008). Prevalence of cognitive impairment without dementia in
the United States. Annals of Internal Medicine, 148(6), 427–434. doi:10.7326/00034819-148-6-200803180-00005
Rasmusson, D. X., Rebok, G. W., Bylsma, F. W., & Brandt, J. (1999). Effects of three types
of memory training in normal elderly. Aging Neuropsychology Cognition, 6, 56–66.
doi:10.1076/anec.6.1.56.790
Recanzone, G. H. (2000). IO cerebral cortical plasticity: Perception and skill acquisition. The
new cognitive neurosciences, 237. ISBN: 0262071959
88
Reisberg, B., Ferris, S., de Leon, M. J., & Crook, T. (1982). The Global Deterioration Scale
for assessment of primary degenerative dementia. American Journal of Psychiatry,
139(9), 1136–1139.
Rensink, R. A., O’Regan, J. K., & Clark, J. J. (1997). To see or not to see: The need for
attention to perceive changes in scenes. Psychological Science, 8(5), 368–373. doi:
10.1111/j.1467-9280.1997.tb00427.x
Reuter-Lorenz, P. A., Jonides, J., Smith, E. E., Hartley, A., Miller, A., Marshuetz, C., &
Koeppe, R. A. (2000). Age differences in the frontal lateralization of verbal and
spatial working memory revealed by PET. Journal of Cognitive Neuroscience, 12(1),
174–187. doi:10.1162/089892900561814
Robin, N., & Holyoak, K. J. (1995). Relational complexity and the functions of prefrontal
cortex. In M. S. Gazzaniga (Ed.), The cognitive neurosciences (pp. 987–997).
Cambridge, MA: MIT Press , xiv, 1447
Rozario, P. A., Kidahashi, M., & DeRienzis, D. R. (2011). Selection, optimization, and
compensation: strategies to maintain, maximize, and generate resources in later life in
the face of chronic illnesses. Journal of Gerontological Social Work, 54(2), 224–239.
doi: 10.1080/01634372.2010.539589
Saczynski, J., Margrett, J. A., & Willis, S. L. (2004a). Older adults’ strategic behavior:
Effects of individual versus collaborative cognitive training. Educational
Gerontology, 30(7), 587–610. doi: 10.1080/03601270490466985
Saczynski, J. S., Willis, S. L., & Schaie, K. W. (2004b). Strategy use and reasoning training.
Aging Neuropsychology and Cognition, 9, 48–60. doi:10.1076/anec.9.1.48.836
89
Salomon, G., Perkins, D. N., & Globerson, T. (1991). Partners in cognition: Extending
human intelligence with intelligent technologies. Educational researcher, 20(3), 2–9.
doi:10.2307/1177234
Salthouse, T. A. (1992). Reasoning and spatial abilities. In F. I. M. Craik & T. A. Salthouse
(Eds.), The handbook of aging and cognition (pp. 167–212). Hillsdale, NJ: Erlbaum.
ISBN-13: 9780805807134
Salthouse, T. A. (1993). Speed mediation of adult age differences in cognition.
Developmental Psychology, 30, 240–259. doi:10.1037/0012-1649.29.4.722
Salthouse, T. A. (1996). The processing-speed theory of adult age differences in cognition.
Psychological Review, 10, 403–428. doi:10.1037//0033-295X.103.3.403
Salthouse, T. A. (2005). Effects of aging on reasoning. In K. J. Holyoak & R. G. Morrison
(Eds.), Cambridge handbook of thinking and reasoning (pp. 589–605). Cambridge,
UK: Cambridge University Press. ISBN: 0521531012 ISBN-13: 9780521531016
Salthouse, T.A (2013). Selectivity of attrition in longitudinal studies of cognitive function.
Journal of Gerontology, Series B: Psychological Sciences and Social Sciences.
doi:10.1093/geronb/gt046
Schaie, K. W. (1996). Intellectual development in adulthood: The Seattle longitudinal study.
New York, NY: Cambridge University Press. ISBN-13: 9780521430142
Schaie, K. W., & Willis, S. L. (1999). Theories of everyday competence and aging. In V. L.
Bengtson & K. W. Schaie (Eds.), Handbook of Theories of Aging (pp. 174-195). New
York, NY: Springer. ISBN-13: 9780826112347
90
Schaie, K. W, Willis, S. L., Hertzog, C., & Schulenberg, J. E. (1987). Effects of cognitive
training on primary mental ability structure. Psychology and Aging, 2, 233–242.
doi:10.1037//0882-7974.2.3.233
Siegler, I. C., Hooker, K., Bosworth, H. B., Elias, M. F., & Spiro, A. (2010). Adult
development, aging, and gerontology. In J. M. Suls, K. W. Davidson, & R. M. Kaplan
(Eds.), Handbook of health psychology and behavioral medicine (pp. 147–162). New
York, NY: Guilford Press. ISBN: 1606238957 ISBN-13: 9781606238950
Springer, M. V., McIntosh, A. R., Winocur, G., & Grady, C. L. (2005). The relation between
brain activity during memory tasks and years of education in young and older adults.
Neuropsychology, 19(2), 181–192. doi:10.1037/0894-4105.19.2.181
Stern, Y. (2002). What is cognitive reserve? Theory and research application of the reserve
concept. Journal of the International Neuropsychological Society, 8, 448–460.
doi:10.1017/S1355617702813248
U.S. Bureau of Labor Statistics & U.S. Census Bureau. (2007) Current population survey.
Retrieved from http://www.bls.census.gov/cps/cpsm ain.htm
U.S. Census Bureau. (2011). The older population: 2010 (Publication no. C2010BR-09).
Retrieved from http://www.census.gov/prod/cen2010/briefs/c2010br-09.pdf
U.S. Department of Health and Human Services (2011). Aging in America: Future
Challenges, Promise and Potential before a Special Committee on Aging for the
United States Senate. Retrieved from
http://www.hhs.gov/asl/testify/2011/12/4173.html
Viskontas, I. V., Holyoak, K. J., & Knowlton, B. J. (2005). Relational integration in older
adults. Thinking & Reasoning, 11(4), 390–410. doi: 10.1080/13546780542000014
91
Wadley, V. G., Crowe, M., Marsiske, M., Cook, S. E., Unverzagt, F. W., Rosenberg, A. L.,
& Rexroth, D. (2007). Changes in everyday function in individuals with
psychometrically defined mild cognitive impairment in the Advanced Cognitive
Training for Independent and Vital Elderly Study. Journal of the American Geriatrics
Society, 55(8), 1192–1198. doi: 10.1111/j.1532-5415.2007.01245.
West, R., & Alain, C. (2000). Age‐related decline in inhibitory control contributes to the
increased Stroop effect observed in older adults. Psychophysiology, 37(2), 179–189.
doi: 10.1111/1469-8986.3720179
Willis, S. L. (1991). Cognition and everyday competence. In K. W. Schaie & M. P. Lawton
(Eds.), Annual review of gerontology and geriatrics (pp. 80–109). New York, NY:
Springer. ISBN-13: 9780826164933
Willis S. L., (1996). Everyday cognitive competence in elderly persons: Conceptual issues
and empirical findings. Gerontologist, 36, 595–601. doi:10.1093/geront/36.5.595
Willis, S. L., Cornelius, S. W., Blow, F. C., & Baltes, P. B. (1983). Training research in
aging: Attentional processes. Journal of Educational Psychology, 75, 257–270.
doi:10.1037/0022-0663.75.2.257
Willis, S. L., Jay, G. M., Diehl, M., & Marsiske, M. (1992). Longitudinal change and
prediction of everyday task competence in the elderly. Research on Aging, 14(1), 68–
91. doi: 10.1177/0164027592141004
Willis, S. L., & Schaie, K. W. (1986). Training the elderly on the ability factors of spatial
orientation and inductive reasoning. Psychology and Aging, 1(3), 239–247.
DOI:10.1037/0882-7974.1.3.239
92
Willis, S. L., Tennstedt, S. L., Marsiske, M. M., Ball, K., Elias, J., Koepke, K. M., . . .
Wright, E. (2006). Long-term effects of cognitive training on everyday functional
outcomes in older adults. Journal of the American Medical Association, 296(23),
2805–2814. doi:10.1001/ jama.296.23.2805